rights / license: research collection in copyright - non ...26200/et… · promoter analysis and...

Research Collection

Doctoral Thesis

Promoter analysis and transcriptional regulation of the CMT1A-disease gene peripheral myelin protein PMP22

Author(s): Maier, Marcel

Publication Date: 2003

Permanent Link: https://doi.org/10.3929/ethz-a-004485643

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For moreinformation please consult the Terms of use.

ETH Library

https://doi.org/10.3929/ethz-a-004485643

http://rightsstatements.org/page/InC-NC/1.0/

https://www.research-collection.ethz.ch

https://www.research-collection.ethz.ch/terms-of-use

Diss. ETH No. 14943

Promoter Analysis and Transcriptional Regulation

of the CMT1A-Disease Gene

Peripheral Myelin Protein PMP22

A dissertation submitted to the

SWISS FEDERAL INSTITUTE OF TECHNOLOGY (ETH) ZURICH

for the degree of

Doctor of Natural Science

presented by

Marcel Maier

Dipl. Natw. ETHZ, Switzerland

Born May 23, 1974

Citizen of Zürich (ZH), Switzerland

Accepted on the recommendation of

Prof. Dr. Ueli Suter, examiner

Prof. Dr. Peter Sonderegger, co-examiner

2003

for Carina and my parents

M

YELIN

G

ENE

R

EGULATION

T

ABLE

OF

C

ONTENTS

3

1 S

UMMARY

....................................................................5

2 Z

USAMMENFASSUNG

....................................................7

3 I

NTRODUCTION

.............................................................9

3.1 Origin and differentiation of Schwann cells................................. 93.2 Genetic defects in myelin proteins are associated with hereditary

peripheral neuropathies .............................................................. 113.3 Transcriptional control of myelin genes..................................... 13

3.3.1 Glial transcription factors and their function ......................................153.3.2 Transcriptional control elements in myelin genes ..............................16

3.4 Aim and motivation of the work ................................................ 25

4 M

YELIN

G

ENE

R

EGULATION

IN

VITRO

........................27

4.1 Aims and Experimental strategies .............................................. 274.2 Promoterdeletion study in MSC80 and NIH 3T3 cells .............. 294.3 Is PMP22 regulated by Sox10 in N2a cells? .............................. 324.4 Screening for transcription factors upregulated upon Krox20

induced myelin gene expression................................................. 33

5 R

EGULATORY

E

LEMENTS

ON

THE

-10/0

KB

PMP22 L

AC

Z T

RANSGENE

...............................................37

5.1 Aims and Experimental strategies .............................................. 375.2 Generation of transgenic mice expressing the reporter lacZ gene

under the control of the PMP22 promoters (-10/0kb PMP22 LacZ)..................................................................................................... 37

5.3 Analysis of the first wave of -10/0kb PMP22 lacZ expression during embryonic development.................................................. 39

5.4 -10/0kb PMP22 lacZ expression is strongly upregulated in postnatal neurons and Schwann cells of peripheral nerves ........ 41

5.5 Transgenic ß-gal expression in sensory and motor neurons of cranial nerves.............................................................................. 45

5.6 Tissue specificity of the -10/0kb PMP22 lacZ transgene........... 465.7 -10/0kb PMP22 lacZ transgene regulation in Schwann cells after

loss of axonal contact and in regeneration. ................................ 485.8 Sciatic nerve of PMP22 mutant animals show reduced ß-gal levels

.................................................................................................... 50

M

YELIN

G

ENE

R

EGULATION

T

ABLE

OF

C

ONTENTS

4

6 PMP22 P

ROMOTER

D

ELETION

A

NALYSIS

IN

VIVO

....53

6.1 Late myelination Schwann cell specific elements (LMSE) reside in the 6 kb DNA fragment upstream of promoter ........... 55

6.2 The LMSE confer Schwann cell specificity to the non-cell type-specific

hsp68

promoter and are functionally independent of core promoter 1 and exon 1A ..................................................... 59

6.3 The 4 kb sequence upstream of exon 2, including promoter 2, contains elements directing expression in sensory neurons ....... 62

7 C

OMPUTATIONAL

A

NALYSIS

OF

C

ONSERVED

P

ROMO

-

TER

E

LEMENTS

AND

R

EPETITIVE

G

ENOMIC

R

EGIONS

65

8 D

ISCUSSION

AND

O

UTLOOK

.......................................69

8.1 PART I: Myelin Gene Regulation

in vitro

................................. 698.2 PART II: Regulatory Elements on the -10/0kb PMP22 lacZ ........

Transgene ................................................................................... 738.3 PART III & IV: PMP22 Promoter Deletion Analysis

in vivo

.... 77

9 EXPERIMENTAL METHODS ................................87

9.1 Generation of reporter constructs ............................................... 879.2 Generation of transgenic animals ............................................... 889.3 ß-gal histochemical analysis....................................................... 899.4 ß-gal solution assay .................................................................... 909.5 Quantitative analysis of PMP22 mRNA levels .......................... 919.6 PMP22-lacZ x PMP22-/- and PMP22-lacZ x Tr mice ............... 929.7 Sciatic nerve transsection and crush........................................... 939.8 Immunocytochemistry of dissociated DRG ............................... 939.9 Cell culture, transfection and reporter assays............................. 949.10 Sequence analysis and determination of potential binding sites 96

10 R

EFERENCES

...............................................................9711 A

CKNOWLEDGEMENTS

.............................................11112 C

URRICULUM

V

ITAE

................................................112

S

UMMARY

- Z

USAMMENFASSUNG

5

1 S

UMMARY

Minor changes in Peripheral Myelin Protein 22 (PMP22) gene dosage have profound

effects on the development and maintenance of peripheral nerves. This is evident from

the genetic disease mechanisms in the hereditary peripheral neuopathy Charcot-Marie-

Tooth disease type 1A (CMT1A) and Hereditary Neuropathy with liability to Pressure

Palsies (HNPP) as well as in transgenic animals with altered PMP22 gene dosage. Thus,

regulation of

PMP22

is a crucial aspect in understanding the function of this protein in

health and disease.

As a first approach to study PMP22 transcriptional regulation, I have generated

transgenic mice containing ten kilobases of 5`-flanking region of the

PMP22

gene,

including the two previously identified alternative promoters, fused to a

lacZ

reporter

gene. I showed that this part of the

PMP22

gene contains the necessary information to

reflect the endogenous expression pattern in peripheral nerves during development,

regeneration, and in mouse models of demyelination due to genetic lesions. Transgene

expression is strongly regulated during myelination, demyelination and remyelination in

Schwann cells, demonstrating the important influence of neuron-Schwann cell

interactions in the regulation of

PMP22

. In addition, the region of the

PMP22

gene

present in this transgene also directs expression in sensory and motor neurons.

These results provided the basis for further analysis of the elements that direct these

specific aspects of temporal and spatial regulation of the

PMP22

gene. To this end, I

subdivided the ten kilobase 5’-flanking region of the PMP22 gene and analyzed different

cis-acting elements as a fusion with either the corresponding PMP22 promoter or a

heterologous hsp68 promoter, both together with a lacZ reporter gene in vivo. This

revealed the existence of two separate elements. The first is a late myelinating Schwann

cell specific element (LMSE) located 5’ to promoter 1 of PMP22. The LMSE is strongly

regulated during myelination and is responsible for specific expression of PMP22 during

the later phase of myelination in Schwann cells. The second element, 2kb 5’ of and

including promoter 2 of PMP22, was active postnatally specifically in sensory neurons.

The activities of these two elements contribute to distinct parts of the endogenous

PMP22 expression.

SUMMARY - ZUSAMMENFASSUNG

6

These in vivo studies were complemented by cell culture experiments analysing PMP22

regulatory elements in a promoter deletion study with transient transfection assays of

PMP22 promoter-driven lacZ reporter constructs. Furthermore, I established a system

with which to further analyse the initiation of myelin gene expression in cell culture.

Combined with a bioinformatics-based determination of conserved regulatory elements

and potential binding sites for transcription factors, these different approaches

contributed to a better understanding of the temporal and spatial regulation of PMP22.

Furthermore, these results provide the basis for further dissection of the molecular basis

responsible for late postnatal expression of PMP22 and the corresponding pathways

converging on the LMSE. These pathways may be important in myelin maintenance or in

demyelinating peripheral neuropathies.


7

2 ZUSAMMENFASSUNG

Aenderungen in der Kopienzahl für das Periphere Myelin Protein 22 (PMP22) haben

erhebliche Auswirkungen auf die Entwicklung und Erhaltung des peripheren

Nervensystems (PNS). Dies wurde offenkundig durch die Erforschung der genetischen

Mechanismen der hereditären Motorischen und Sensiblen Neuropathie Charcot-Marie-

Tooth Typ 1A (CMT1A), der Hereditären Neuropathie mit Neigung zu Druckparesen

(HNPP-Hereditary Neuropathy with liability to Pressure Pulsies) sowie dem Studium

von Tiermodellen mit veränderter PMP22 Kopienzahl. Diese Arbeiten zeigten, dass die

Regulation des PMP22 Genes einen essentiellen Aspekt darstellt, um die Rolle dieses

Proteins in der normalen Entwicklung sowie in der Entstehung von Neuropathien zu

verstehen.

Um die transkriptionellen Regulation von PMP22 zu analysieren, generierte ich zunächst

eine Maus, die als Transgen die 10 Kilobasen (kb) der 5’ Region des PMP22 Genes trägt.

Diese 10 kb grosse Region beinhaltet die beiden bekannten PMP22 Promoteren und ist

im Transgen mit einem LacZ Reportergen fusioniert. In der folgenden Analyse zeigte

ich, dass dieser Teil des PMP22 Gens die notwendigen Informationen enthält, um die

endogene Expression im PNS während der Entwicklung, der Regeneration und in

Tiermodellen für Periphere Neuropathien wiederzuspiegeln. Während der

Myelinisierung, der De- und Remyelinisierung wurde die Transgenexpression in

Schwann’schen Zellen, ebenso wie die Expression des endogenen PMP22, stark

reguliert. Dies zeigt, wie wichtig die Interaktion von Neuronen mit den Schwann’schen

Zell für die Regulation von PMP22 ist. Diese 10 kb grosse Region des PMP22 Gens

führt zudem zur Expression des Reportergens in sensorischen und motorischen

Neuronen des PNS.

Diese Resultate dienten als Basis für eine weitere Analyse der Elemente, die für die

spezifische, zeitliche und räumliche PMP22 Expression verantwortlich sind. Zu diesem

Zweck unterteilte ich die 10 kb der 5’ Region von PMP22. Die verschiedenen daraus

resultierenden cis-wirkenden Elemente analysierte ich als Fusion mit den

entsprechenden PMP22 Promotoren respektive mit dem heterologen hsp68 Promotor in

vivo. Als Reportergen wurde weiterhin das LacZ verwendet. Dies führte zur Entdeckung


8

von zwei verschiedenen Elementen. Das erste Element nannten wir das “späte

Myelinisierung- und Schwann’sche Zell-spezifisches” Element (LSME - late

myelinating Schwann cell specific element) welches 5’ vom Promoter 1 lokalisiert

wurde. Während der Myelinisierung ist das LSME verantwortlich für die starke und

spezifische Regulation der PMP22 Expression während der späten Phase der

Myelinisierung. Das zweite Element, lokalisierte ich zwei Kilobasen 5’ vom Promoter 2.

Es war zusammen mit Promotor 2 spezifisch in sensorischen Neuronen postnatal aktiv.

Die Aktivität dieser beiden Elemente reguliert offenbar verschiedene Aspekte der

endogenen PMP22 Expression.

Die in vivo Studien wurden ergänzt durch Zellkulturexperimente. Hier analysierte ich die

regulatorischen PMP22 Elemente in einer Promoterdeletionsstudie nach transienten

Transfektionen von PMP22 Promoter-lacZ Reporterkonstrukten. Zudem etablierte ich

ein System, mit welchem die Analyse der Initiation der Myelingenexpression in Zell-

kultur studiert werden kann. Die Resultate dieser verschieden Ansätze konnten mit

Bioinformatik-basierten Bestimmungen der konservierten, regulatorischen Elemente

sowie der potentiellen Bindungsstellen für Transkriptionsfaktoren kombiniert werden.

Diese verschiedenen Analysen bilden die Basis für das weitere Studium der molekularen

Mechanismen der Expression von PMP22 während der späten Phase der Myelinisierung

und der entsprechenden Signalwege an diesem LSME. Weitere Analysen werden zeigen,

ob die Signalwege an den indentifizierten LSME für die Erhaltung der Myelinschicht

oder bei der Pathogenese von demyelinisierenden, peripheren Neuropathien essentiell

sind.

MYELIN GENE REGULATION INTRODUCTION

9

3 INTRODUCTION

3.1 Origin and differentiation of Schwann cells

The trunk neural crest gives rise to melanocytes, neurones and peripheral glia. Among

the glia are the Schwann cell precursors, found in rat peripheral nerves at E14 to E15,

and in mouse nerves at E12 to E13. In a relatively abrupt transition between E16 and E17

(rat) or E14 and E15 (mouse), immature Schwann cells are formed while migrating into

the periphery along the axonal tracts. These then diverge around the time of birth to give

rise to myelinating and non-myelinating Schwann cells (Fig. 3-1). Here, their fate is

tightly controlled by axonal signals, e.g. by trophic factors such as neuregulin that

promote the survival of the Schwann cells and their precursors (for review see Jessen and

Mirsky, 1999; Lobsiger et al., 2002). Prospectively myelinating Schwann cells form a

1:1 relationship with axons prior to myelination. Schwann cells with this morphology are

often referred to as pro-myelinating Schwann cells. Larger calibre axons (1µm and more)

will be myelinated with each myelinating Schwann cell providing one myelin internode

(Fig. 3-2). The transition from the promyelinating to the myelinating stage is then

accompanied by a number of significant changes in the pattern of gene expression,

including the activation of a set of genes encoding myelin structural proteins and lipid

biosynthetic enzymes, and the inactivation of a set of genes expressed only in immature

or nonmyelinating Schwann cells. Smaller calibre axons remain unmyelinated and are

ensheathed in bundles of 5-30 axons by the non-myelinating Schwann cells (Friede and

Samorajski, 1968). Both types of Schwann cells are surrounded by a basement

membrane, which separates them from other components of the endoneurium in the

peripheral nerve and anchors them in the extracellular matrix.

The myelin sheath formed by Schwann cells can be of extremely large diameter,

consisting of up to 100 wrappings. Schwann cell myelin contains incisure (Schmidt-

Lantermann incisures) which traverse the compact myelin. These funnel-shaped domains

of uncompacted myelin contain remnants of cytoplasm and, similar to the paranodal

loops, contain different proteins than are found in compact myelin (for review see

Scherer, 1999; Scherer and Arroyo, 2002).


10

Figure 3-1: Diagrammatic representation of the development of Schwann cells inrat and mouse with their molecular markers (adapted with modifications fromJessen and Mirsky, 1999).

Upon nerve transection, or when dissociated and placed in tissue culture, myelinating

and non-myelinating Schwann cells undergo a reversion in molecular phenotype

comparable to that seen in immature nerves. In regenerating nerves, these Schwann cells

again differentiate into myelinating and non-myelinating Schwann cells. Thus the

Schwann cell molecular and morphological phenotype is reversible from the immature

Schwann cell state onwards.

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

!�

� ��

��

��

��

��

��

��

�� " � ��

#$ #� #� ��


11

Figure 3-2: Simplified and schematic anatomy of the spinal nerves of the peripheralnervous system. Abbreviations: DRG: Dorsal root ganglion, NMJ: Neuromuscularjunction

3.2 Genetic defects in myelin proteins are associated withhereditary peripheral neuropathies

The hereditary peripheral neuropathy Charcot-Marie-Tooth disease (CMT) comprises a

heterogeneous group of genetic human disorders that affect the peripheral nervous

system (PNS) with an estimated prevalence of 1:2500 (Skre, 1974). In recent years

mutations in up to ten genes have now been identified in humans as culprits in different

forms of CMT (reviewed by Berger et al., 2002; Maier et al., 2002b; Young and Suter,

2001), suggesting that there are different important players in the interplay between

Schwann cells and neurons. Three of the PNS myelin proteins, PMP22, Cx32 and MPZ,

have been linked to one of the most common forms of hereditary motor and sensory

neuropathies (HMSN), the Charcot-Marie-Tooth disease type 1 (CMT1; reviewed by

Suter et al., 1995).

In my thesis project, I focused on the gene encoding the peripheral myelin protein 22

(PMP22) which is the gene involved in the vast majority of patients affected by the


12

hereditary peripheral neuropathy Charcot-Marie-Tooth disease (CMT; subtype CMT1A;

reviewed in Maier et al., 2002b; Young and Suter, 2001; Suter and Snipes, 1995).

Patients suffering from CMT1A, the most common form, show slowed nerve conduction

velocities, reduced compound motor and sensory nerve action potentials, progressive

distal limb weakness, sensory loss, and decreased reflexes (Dyck et al., 1993).

Histologically, characteristic findings in these patients are progressive demyelination of

motor and sensory nerves associated with incomplete remyelination. In the majority of

the patients the autosomal-dominant CMT1A is due to an intrachromosomal duplication

on chromosome 17p11.2, leading to the presence of an extra copy of the gene for

peripheral myelin protein 22. The reciprocal deletion of the same DNA fragment has also

been identified in patients and is associated with hereditary neuropathy with liability to

pressure palsies (HNPP; Chance et al., 1993), which is usually not progressive but rather

characterized by temporary palsies after pressure trauma (Windebank, 1993; Amato et

al., 1996). In addition, PMP22 missense mutations have been found in some rare cases of

familial CMT1A without the usual chromosomal duplication (Roa et al., 1993; Naef and

Suter, 1999).

The findings derived from the genetics of CMT1A and HNPP, and the fact that the

PMP22 gene is neither disrupted by the CMT1A duplication nor mutated in these

patients, led to the suggestion that altered PMP22 gene dosage may be responsible for

these two frequent neuropathies. The observation that homozygous CMT1A duplication

patients are often more affected than heterozygous relatives supports this hypothesis

(Lupski et al., 1991; Kaku et al., 1993). Additional evidence that PMP22 is a dosage-

sensitive gene has been provided by accurate animal models in which mice and rats carry

variable numbers of the PMP22 gene (Adlkofer et al., 1995; Huxley et al., 1996; Magyar

et al., 1996; Sereda et al., 1996). In particular, a twofold increased PMP22 gene dosage

in transgenic rats led to a morphological phenotype comparable to CMT1A. Transgenic

mice retaining only one functional PMP22 allele develop, as expected from the

analogous human disease, a pathology comparable to HNPP, whereas mice lacking

PMP22 completely develop a demyelinating peripheral neuropathy reminiscent of severe

CMT1A (Adlkofer et al., 1995). Two different missense mutations in the PMP22 gene

have also been found in the allelic mouse mutants Trembler (Tr) and Trembler-J (Tr-J),

which initially implicated PMP22 as the critical gene within the CMT1A duplication.


13

Since these animals show a PNS-specific dismyelination, they have been proposed to be

suitable models for CMT1A (Suter et al., 1992a, b).

Taken together, these findings suggest that the PMP22 gene must be very tightly

regulated since already slightly higher or lower gene dosage (150% in CMT1A; 50% in

HNPP) lead to a disease phenotype.

Interestingly, a comparable phenomenon is observed for the proteolipid protein (Plp)

gene. Similar to the situation in CMT1A for PMP22 increased gene dosage resulting

from a duplication of the proteolipid protein (Plp) gene is one of the causes of the

Pelizaeus-Merzbacher disease (PMD) and leads to myelination abnormalities in the CNS

(reviewed by Anderson et al., 1999; Yool et al., 2000). Transgenic mice carrying extra

copies of the wild-type Plp gene provide a valid model of PMD. Variations in gene

dosage can cause a wide range of phenotypes from severe, lethal dysmyelination through

late-onset demyelination.

3.3 Transcriptional control of myelin genes

The extremely high and transient demand for newly synthesized myelin proteins during

nerve development and regeneration is accompanied by the coordinated expression of

genes which encode myelin components (reviewed by Toews et al., 1997). Deciphering

the molecular basis of this regulation is crucial for our understanding of the interplay

between neurons and glia cells in myelin formation in health and in diseases such as

multiple sclerosis and neuropathies.

In the PNS, two main strategies have been employed to address this issue. First, efforts

have been made to identify potential master regulators of myelin gene expression in

analogy to the transcription factor MyoD in muscle development (reviewed by Borycki

and Emerson, 1997). This search has proven to be quite difficult, although during the last

years, some candidates have been identified. These will be introduced in chapter 2.3.1

below.

The second approach involves the search for regulatory regions within myelin genes,

since cell-type specific transcriptional control mechanisms can be unravelled by the


14

identification and characterization of cis-acting control elements in genes preferentially

expressed in a given cell-type or tissue. A number of myelin genes have been studied

with regard to their transcriptional regulation. Sometimes this was done in transgenic

mice, more frequently in tissue culture systems by transient transfection and in vitro by

DNA binding studies. Each of these methods used in gene regulation studies has its

advantages and its limitations. Transgenic approaches, for instance, unequivocally

determine the capacity of a regulatory region to drive temporal and cell type-specific

expression. Therefore transgenic mice provide an excellent assay system to examine

gene regulation, in particular under conditions which require numerous developmental

and physiological signals for correct interactions (reviewed by Duchala et al., 1996).

This situation is obvious in the intensive axon-glia interactions in PNS development and

thus, regulation of myelin genes and their regulators are most accurately characterized in

vivo. Such studies have the disadvantage that they are time-consuming and labor-

intensive, and hardly suited for an in-depth fine-mapping analysis. By contrast, transient

transfections in tissue culture combined with in vitro DNA-binding experiments provide

an efficient tool for a rapid molecular dissection of a regulatory region (for example, see

Brown and Lemke, 1997). However, they do not always give a clear answer as to whether

the region under study is by itself sufficient to elicit the cell type-specific expression of

the gene to which it belongs. Furthermore such studies require a system in which the

gene of interest is expressed at reasonable levels compared to the expression in vivo

(compare also chapter 4.1 of this thesis). Especially when studying myelin genes there

are very often limitations due to the fact that Schwann cells do not myelinate in vitro

without co-culturing with neurons. Taking these limitations into consideration, a

combination of in vivo, tissue culture, and in vitro methods is likely to give the most

meaningful results. The region responsible for temporal and tissue-specific expression

can first be localized to a larger genomic fragment by transgenic techniques. Well-chosen

tissue culture systems and DNA-binding studies can then be used to map the potentially

important cis-acting elements within this region. Finally, the importance of these

elements should again be confirmed in vivo, for example by deleting the regulatory

element by homologous recombination in embryonic stem cells, which may result in the

generation of a cell-type specific allele or a cell type-specific null mutant in case of a cell

type- specific enhancer (for example see Ghazvini et al., 2002). In the past, some myelin


15

genes and their regulatory elements have been analyzed in separate studies both in

transgenic mice and in tissue culture. Some aspects of the transcription regulation of

myelin genes with the focus on those which are expressed during myelination in the PNS

are introduced below in chapter 3.3.2. They will give an idea about the variety of

potential regulatory mechanisms which might be involved in the transcriptional

regulation for PMP22.

3.3.1 Glial transcription factors and their function

Several transcription factors that exert a pivotal role in Schwann cells have been

identified as components of the molecular mechanisms initiating myelination. These

include the zinc finger protein Krox20/Egr-2 and the POU protein Oct-6/SCIP/Tst-1/

Pou3f1 (referred to Oct-6 in the thesis; reviewed by Zorick and Lemke, 1996; Zorick et al.,

1999). In support of this hypothesis, the generation of null mutant mice has revealed

important roles for these factors in the process of myelination. In both Krox20 and Oct-6

mutants, Schwann cells appear morphologically arrested at the promyelinating stage

(Bermingham et al., 1996; Jaegle et al., 1996; Topilko et al., 1994). Detailed analysis of

the Krox20 mutant showed severe defects in Schwann cell development resulting in an

hyperproliferation and presumed differentiation arrest at the promyelinating stage

(Topilko et al., 1994). Although Oct-6 mutant Schwann cells exhibit a phenotype similar

to that of Krox20 mutants during the first week after birth, myelination subsequently

resumes (Jaegle et al., 1996). This has led to the suggestion that Oct-6 is involved in the

timing of myelination, whereas Krox20 would be integral to the myelination program

(Jaegle and Meijer, 1998). Consistent with the latter hypothesis, genome expression

profiling studies in Schwann cells revealed that Krox20 regulates multiple genes

involved in myelin formation (Nagarajan et al., 2001). These results are reinforced by

clinical studies on patients with hypomyelinating neuropathies including congenital

neuropathies, Charcot-Marie-Tooth type 1, and Dejerine-Sottas syndrome (compare

chapter 3.2). Some of these patients carry dominant or recessive point mutations

affecting different domains of the Krox20 protein (Boerkoel et al., 2001; Timmerman et

al., 1999; Warner et al., 1998).


16

Additional transcription factors that may affect late steps in the differentiation of

Schwann cells and may be involved in myelin gene regulation include Pax-3, c-Jun,

Krox24/Egr-1 and Sox10 (reviewed by Scherer, 1997; Wegner, 2000a). Sox10 has been

identified as a common transcriptional modulator of Oct-6, Krox20 and Pax3 potentially

conferring cell specificity to interacting transcription factors in developing and mature

glia (Kuhlbrodt et al., 1998). Indeed, recent results suggest a crucial role of Sox10 in

peripheral glia development (Britsch et al., 2001; Paratore et al., 2001) and regulation of

the major PNS myelin protein MPZ by Sox10 has been demonstrated (Peirano et al.,

2000) (compare chapter 4.3).

Recently, it was shown that the POU domain transcription factor Brn-5/Pou6f1 has a

developmental expression pattern inverse to that of Oct-6, with Brn-5 stably expressed in

the adult myelinating Schwann cell, but virtually absent during promyelination (Wu et

al., 2001). Beside Brn-5, the discovery that another POU domain transcription factor

Brn-2/Pou3f2 is not restricted to CNS remyelination, but is expressed in a pattern similar

to that exhibited by Oct-6 during Schwann cell myelination of neonatal nerves and

during regeneration of crushed adult nerve, made this transcription factor to an additional

candidate for the regulation of myelin genes (Sim et al., 2002).

Besides the Schwann cell-specific transcription factors mentioned above binding sites

have been also identified for various groups of transcription factors such as AP-1, CREB,

STAT, NF-κB or several members of the nuclear receptor family, which are known to

translate extracellular signals used by many different cell-types into changes of gene

expression. These are well-described components of trans-acting factors in the

transcription machinery of myelinating glia and may also take part in the regulation of

myelin gene expression in the PNS (reviewed by Wegner, 2000a).

3.3.2 Transcriptional control elements in myelin genes

Much biological regulation takes place at the level of transcription initiation, where an

assortment of varied ‘enhancer’ and ‘silencer’ sequences serve as docking sites for

transcriptional activators and repressors, the precise combination of which controls gene


17

expression. Although inspection of myelin gene sequences has suggested the presence of many

transcription factor binding sites, so far only a few transcription factors have been documented to

bind to the promoter regions of myelin genes in vitro or in vivo. Nevertheless, the focus of the

following chapter is on myelin gene transcription in the PNS, in particular on the structure and

regulation of myelin gene promoters. An overview of regulation of glial transcription

with the corresponding references can be found in Tab 1. The topic is reviewed in more

detail in (Wegner, 2000a, b).

Myelin Protein Zero (MPZ/P0)

One of the best-analyzed myelin genes is the myelin protein zero gene (MPZ/P0), which

encodes a tetramer-forming 31-kilodalton (kDa) transmembrane glycoprotein of the

immunoglobulin superfamily with specific expression in the Schwann cell lineage. MPZ

expression starts early in these cells, possibly in neural crest cells already committed to a

glial fate (Hagedorn et al., 1999; Lee et al., 1997). It is massively upregulated in

Schwann cells during the onset of myelination. MPZ is a major myelin component

accounting for more than 50% of the protein in PNS myelin. It is actively involved in

stabilizing the myelin sheath by binding to other MPZ molecules in opposing

membranes via homophilic interactions (for review see Mirsky and Jessen, 1999;

Scherer, 1997).

Experiments on the 5’ flanking region of the rat MPZ gene have shown, both in culture

and in transgenic mice, that 1.1 kilobases (1kb) upstream of the transcription start site are

sufficient to drive expression in Schwann cells (Lemke et al., 1988; Messing et al.,

1992). However, transgene expression with this construct in mice was variable between

single Schwann cells and there was evidence of ectopic expression. Consistently, another

group could not express lacZ under control of a similar small pieces of the MPZ

promoter specifically in Schwann cells. A more consistent expression was obtained using

the complete mouse MPZ gene, including 6kb of 5’ flanking sequences plus all exons

and introns (Feltri et al., 1999). Nevertheless, the 1.1kb part of the MPZ promoter is

capable of recapitulating some essential features of MPZ expression, including low-level

expression during early phases of PNS development and increased expression in

myelinating Schwann cells. Mapping studies in Schwann cell cultures have been used to

dissect this 5’ flanking region into functional domains (Brown and Lemke, 1997). These


18

include (1) a core promoter, which has little activity on its own and which spans

positions -100 to +45 relative to the transcription start site (defined as +1); (2) a potently

activating proximal promoter region from position -350 to -100, which exhibits high

sequence conservation among rat, mouse and humans, and which is responsible for

greater than 50% of the activity of the MPZ promoter in Schwann cell cultures; and (3)

an accessory distal region from positions -910 to -315 which makes up for the rest of the

activity (Table 1).

Similarly structured promoters are also found in other myelin genes, for instances the P2

gene (Bharucha et al., 1993) and the myelin-associated glycoprotein gene (MAG)

(Grubinska et al., 1994; Laszkiewicz et al., 1997; Ye et al., 1994) (Table 1).

DNase footprinting of the MPZ promoter revealed the presence of numerous binding

sites in all three parts (Brown and Lemke, 1997). Some binding sites were occupied by

proteins from glial and nonglial cells, while others showed differential protection in one

cell type only. Most of the identified cis-acting elements in these experiments do not

correspond to known transcription factor binding sites, with the exception of a GC rich

binding site and CAAT boxes in the core promoter region. The GC-rich binding site is

recognized by Sp1 family proteins, and the CAAT boxes are recognized by NF-Y

(Brown and Lemke, 1997). Both CAAT boxes and GC-rich binding sites are found in a

number of other glial promoters (Table 1), including that of Oct-6 (Kuhn et al., 1991)

and Promoter 1 (see section below) of PMP22 (Suter et al., 1994),

Myelin Basic Protein (MBP)

Myelin basic proteins (MBP) occur both in myelinating Schwann cells and in

oligodendrocytes as multiple isoforms transcribed from the same gene. MBP are highly

charged proteins with high affinity for membranes and mediate the close apposition of

myelin membranes on their cytoplasmic sides. MBP makes up 3% of total myelin in the

PNS and 30% of total myelin protein in the CNS. The MBP gene is part of the larger

Golli-mbp gene, which is conserved between humans and rodents (Campagnoni et al.,

1993; Pribyl et al., 1993). Numerous transgene experiments have been carried out with

5’flanking regions, employing varying length ranging from 256bp to 6.5kb taken from

mouse or human MBP gene (reviewed by Ikenaka and Kagawa, 1995). Interestingly, all

the transgenes exhibited expression in cells of the oligodendrocyte lineage, although the


19

penetrance of transgene expression was lower for example in the transgenes containing

only 256bp (Goujet-Zalc et al., 1993). None showed expression in Schwann cells.

Nevertheless, this short proximal promoter region seems to contain the necessary

information for oligodendrocyte-specific expression with further upstream sequences

amplifying the effect. Expression of the transgenes comes on in the oligodendrocyte

lineage with the developmental profile typical of MBP. Most transgenes, however,

exhibit a decline in expression after peaking during the active myelination period, and do

not show the strong maintained level of expression seen for MBP during adulthood.

Comparable observations were also made with a transgene under the control of the

immediate 5’ flanking region of the CNP gene where the correct temporal expression of

the transgene was obtained only in oligodendrocytes (Gravel et al., 1998) (Table 1).

Together with the MBP transgenes, it was proposed that differential regulatory elements

may exist for oligodendrocyte-specific versus Schwann cell-specific expression and for

early versus late expression of myelin genes.

Indeed, a recent in vivo promoter study identified DNA elements responsible for MBP

expression in the PNS (Forghani et al., 2001). Evidence was provided for the

participation of multiple, widely distributed, positive and negative elements in the overall

control of MBP expression. Especially, all constructs bearing a 0.6 kb far- upstream

sequence, designated Schwann cell enhancer 1 (SCE1), expressed at high levels in

myelin-forming Schwann cells. In addition the 0.6 kb SCE1 alone shows robust targeting

activity independent of other MBP 5' flanking sequence.

The MBP promoter has also been extensively characterized in transient transfections and

in vitro using transcriptionally competent brain extracts. Taken together these studies

agreed with the in vivo analysis in that a comparatively short region preceding the

transcription start site was found to contain the information necessary for

oligodendrocyte-specific expression (Table 1). Using various methods to detect protein-

DNA interactions in vitro, this region was shown to contain numerous sites capable of

binding both ubiquitous and cell type-specific nuclear proteins (e.g. Devine-Beach et al.,

1990). Most of these binding activities are still poorly characterized at the molecular

level. However, it has been shown that this proximal promoter region contains response

elements for the thyroid hormone receptor (TR) (Farsetti et al., 1992; 1991) and for

members of the NF-I and Sp1 family of transcription factors (Aoyama et al., 1990;


20

Tamura et al., 1988) (Table 1). The latter two transcription factors are fairly ubiquitous

proteins and are generally believed to contribute to basal rather than to cell type-

dependent expression. Nevertheless, also an ubiquitous transcription factor like NF-I

may contribute to oligodendrocyte specificity by additional protein-protein and DNA-

protein interactions around the binding site of NF-I. Indeed, there are indications that the

NF-I site in the MBP promoter is part of a larger composite regulatory element, which is

flanked in the case of the human promoter on its 5’ site by a site for a not further

caracterized activity called MEBA (Taveggia et al., 1998). In the mouse promoter the 3’

flanking region is assumed to mediate binding of an activity called M1 (Aoyama et al.,

1990). Thus, one might speculate that such an NF-I centered nucleoprotein complex

could be an important determinant in creating the oligodendrocyte specificity of MBP

gene expression, without NF-I itself being oligodendrocyte specific.

TABLE 1: Regulatory regions of genes expressed in myelinating glia

(adapted with modifications from Wegner, 2000a)

Gene Expression Regulatory regions Binding sites Transgene (expression) Reference

P0 PNS Core: - 100 to + 45 TATA, CAAT, GC,Sox10, other fp

1.1 kb 5' flank (SC) Lemke et al., 1988;Prox: - 350 to - 100 Messing et al., 1992;Dist: - 915 to - 350 Brown et al., 1997

P2 PNS Core: - 293 to + 125 TATA, CAAT – Bharucha et al., 1993Prox: - 435 to - 293Dist: - 870 to - 435

PMP-22 PNS P1: - 298 to + 172 NFI, CAAT, TATA 10 kb 5' flanking (myelinating SC)

Suter et al., 1994ubiq. P2: - 352 to + 144 GC

L1 PNS, CNS Core: - 150 to + 118 NFI, GC, K-PaxhomeoproteinsNRSF

2.9 kb 5 Kallunki et al., 1997,1998;Enhancer: int 1 ex 1±4, int 1±3

Silencer: int 2 (CNS, PNS) Meech et al., 1999

MBP CNS, PNS Core: - 36 to + 12 TATA, GC, NFI, MEBA, 6.5±0.3 kb 59 e.g., Goujet-Zalc etal., 1993;Prox: - 256 to - 36 M1, TR, other FP Øank (OL, not SC)

FynRE: - 675 to - 647 STAT, C/EBP Tamura et al.,1990;Umemori et al., 1999

PLP CNS, (PNS) Core: - 186 to 1 87Promoter: - 1038 to

+ 87Enhancer: int 1

TATA, CAAT, MyT1,other fp

human: 4.2 kb 59 and1.5 kb 3' flank;mouse: 2.4 kb 5'flank, ex. 1,2, int 1(OL)

Nave and Lemke,1991; Berndt et al.,1992; Nadon et al.,1994; Wight et al.,1993, 1997

CNP CNS, PNS P1 TATA 4-kb 5' flank Gravel et al., 1998ubiq. P2 TATA (OL, early SC)

MAG CNS, PNS Core: - 138 to 1 21 GC, AP2 – Ye et al., 1994;Prox: - 583 to - 138 Grubinska et al.,

1994Dist: - 861 to - 583

Krox-20 PNS Immature SC element (immature SC, 5' flank)

SRE, CArG-1, CRE,AP-1, SP-1

Oct-6/Tst-1 PNS, CNS Core: - 93 + 49 TATA, CAAT, GC SC Enhancer, about 5kb 3' flank

Kuhn et al., 1991;Enhancer: - 5kb AP-1, ER Renner et al., 1996a

ubiq., ubiquitous expression; core, core promoter; prox, proximal promoter; dist, distal promoter; fp, footprinted region; SC, Schwann cell; OL, oligodendrocyte; ex,exon; int, intron.

Maier et al., 2002

' flank

review: Beckmann et al., 1997

Myelinating SC element (myelinating SC, 3' flank)

Ghislain et al., 2002

Mandemakers et al., 2000Ghazvini et al., 2002


21

Proteolipid Protein (PLP)

The promoter topology of the gene encoding the proteolipid protein (PLP) is quite

different from those described so far. PLP is a highly conserved myelin protein with four

transmembrane domains, and functions in oligodendrocyte development, myelin

compaction, and axonal integrity. It is expressed as the two splice-variants, PLP and DM-

20, which exhibit different developmental expression profiles and display non-redundant

functions (Nadon et al., 1994, and references therein). The major splice variant PLP

makes up 40% of total myelin protein in the CNS. PLP is also expressed in Schwann

cells, albeit at much lower levels (e.g. Garbern et al., 1997).

For the rat, mouse and human PLP promoters the proximal 186, 145, and 204 base pairs

(bp), respectively, have been described as sufficient to generate basal promoter activity in

transiently transfected cells (Berndt et al., 1992; Nave and Lemke, 1991; Wight et al.,

1993). This region appears to contain multiple start sites that are differentially used in

oligodendrocytes versus Schwann cells (Kamholz et al., 1992; Scherer et al., 1992). This

region as well as the sequences immediately upstream contain a number of protein

binding sites, as judged by DNase footprinting experiments and electrophoretic mobility

shift assays (EMSA) (Berndt et al., 1992; Nave and Lemke, 1991). However, even 1kb

immediately upstream of the transcription start site is not sufficient to direct

oligodendrocyte-specific expression of a transgene (Nadon et al., 1994). Instead, 4.2kb

of 5’flanking sequence from the human PLP gene were needed in combination with

1.5kb of 3’ flanking sequence to obtain oligodendrocyte-specific expression in transgenic

animals. Several additional transgenes with oligodendrocyte-specific expression were

generated using different regions from the mouse PLP gene, leading also to PNS specific

expression (reviewed by Ikenaka and Kagawa, 1995; Kagawa et al., 1994; Readhead et

al., 1994; Wight et al., 1993). One conclusion to draw from the sum of transgenic

experiments is to postulate the presence of multiple oligodendrocyte-specific regulatory

elements in the PLP gene, one in the 5’ flanking region and the other in intron 1 - at least

in the mouse sequence.

To assume the presence of regulatory element necessary for oligodendrocyte-specific

expression within intron 1 is not unreasonable, as important regulatory elements have

been found in intronic sequences of many genes. In the case of L1, for instance,

regulatory elements have been found both in intron 1 and in intron 2, each responsible


22

for a specific expression pattern (Kallunki et al., 1998; 1997; Meech et al., 1999). The L1

gene codes for an integral membrane protein of the immunoglobulin superfamily,

modulates neuron-neuron and neuron-glia interactions, and exhibits strong expression in

the PNS during embryonic development and later in non-myelinating Schwann cells.

Oct-6/SCIP

In the case of the Oct-6 gene regulatory elements were mapped as DNase I-

hypersensitive sites (HSS). A combination of two of the HSS was shown to act as Oct-6

Schwann cell-specific enhancer (SCE) and were located very distally to the gene

(Mandemakers et al., 2000). The SCE is sufficient to drive spatially and temporally

correct expression of Oct-6, during both normal peripheral nerve development and

regeneration. Because Oct-6 expression is under the control of axonal signals during

nerve development and regeneration, this SCE provides a cis-acting genetic element that

responds to converging signalling pathways to drive myelination in the PNS.

Consequently in a recent study an Oct-6 allele with reduced expression in Schwann cells

was generated through deletion of the SCE in the Oct-6 locus (Ghazvini et al., 2002).

Indeed, the analysis of mice homozygous for this allele reveals that rate-limiting levels of

Oct-6 in Schwann cells are dependent on the SCE and that this element does not

contribute detectably to Oct-6 regulation in other cell types.

Krox20/Egr-2

Analysis of cis-acting elements governing Krox20 expression with transgenes in

Schwann cells revealed the existence of two separate elements (Ghislain et al., 2002).

The first, designated immature Schwann cell element (ISE), was active in immature but

not myelinating SC, whereas the second, designated myelinating Schwann cell element

(MSE), was active from the onset of myelination through adulthood in myelinating SC.

In vivo sciatic nerve regeneration experiments demonstrated that both elements were

activated during this process, in an axon-dependent manner. Together the activity of

these elements reproduced the profile of Krox20 expression during development and

regeneration. The MSE was localised to a 1.3 kb fragment 35 kb downstream of Krox20.

The identification of multiple Oct-6 binding sites within this fragment suggested that

Oct-6 directly controls Krox20 transcription. The ISE was located in the -31/-4.5kb


23

region of the Krox20 gene. A series of four different overlapping fragments from this

region were tested in mouse transgenesis by fusion to a ß-globin minimal promoter/lacZ

reporter. Although regulatory sequences controlling Krox20 in the nerve roots were

identified, none of the constructs showed expression distally in the peripheral nerves.

Taken together, these data indicate that, although Krox20 is expressed continuously from

15.5 dpc in Schwann cells, the regulation of its expression is a biphasic, axon-dependent

process involving two cis-acting elements that act in succession during development. In

addition, they provide insight into the complexity of the transcription factor regulatory

network controlling myelination.

In vitro studies showed that the Krox-20 promoter contains two sequences similar to

serum response elements, with an CArG-box as an inner core element (CArG-1 and

CArG-2). CArG-1 is responsible for the serum and AP-1 responsiveness of Krox-20 via

PKC-dependent and -independent pathways, respectively (reviewed by Beckmann and

Wilce, 1997).

Peripheral Myelin Protein 22 (PMP22)

The 22 kDa hydrophobic glycoprotein PMP22 is predominantly expressed by

myelinating Schwann cells in peripheral nerves, where the PMP22 protein is

incorporated into compact myelin (Haney et al., 1996; Snipes et al., 1992; Welcher et al.,

1991). It accounts for 2-5% of the total protein found in isolated myelin preparations

(Pareek et al., 1993). The PMP22 molecule consists of an 18kDa polypeptide core, as

predicted from its primary amino acid sequence, and of carbohydrates, which are linked

to an asparagine residue. Some of the PMP22-bound glycosyl moieties carry the L2/

HNK-1 carbohydrate epitope, which has been implicated in intercellular recognition and

adhesion processes (Snipes et al., 1993).

Although PMP22 is found mainly in Schwann cells and seems to be required for correct

development of peripheral nerves, the maintenance of axons and the determination of

myelin thickness and stability (Adlkofer et al., 1997a; 1995), it is not myelin-specific.

Hence a more general role of PMP22 in cell biology has been proposed. Indeed, the

cDNA encoding PMP22 was first isolated from NIH3T3 fibroblasts as a growth arrest-

specific gene (gas-3) (Manfioletti et al., 1990; Schneider et al., 1988). The function of

PMP22 is unknown, but results from in vitro studies suggest that exogenously altered


24

PMP22 expression affects proliferation, cell shape, and spreading, as well as apoptotic

cell death of Schwann cells (Brancolini et al., 1999; 2000; Fabbretti et al., 1995; Zoidl et

al., 1995). On the other hand, studies of certain PMP22 mutations with an aberrant

intracellular trafficking (Naef and Suter, 1999; 1998) and association of PMP22 with

Calnexin (Dickson et al., 2002) propose a role of PMP22 in membrane organization in

the mechanism of myelin formation.

In the rat, two different PMP22 cDNAs, SR13 and CD25, have been isolated (Spreyer et

al., 1991; Welcher et al., 1991), and meanwhile analogous transcripts been described in

humans and mice (Suter et al., 1994; van de Wetering et al., 1999). These transcripts

consist of different 5’ untranslated regions, resulting from two alternatively transcribed

exons (Suter et al., 1994). The mapping of separate transcription start sites to each of

these exons indicated that PMP22 expression is regulated by two alternatively used

promoters. The relative expression of the alternative PMP22-transcripts is tissue specific,

and high levels of the exon 1A-containing transcript (CD25, 1A-PMP22) are coupled to

myelin formation. In contrast, transcripts containing exon 1B (SR13, 1B-PMP22) are

also present in non-neural tissues such as the lung, intestine, heart and skeletal muscle

(Suter et al., 1994). In addition, PMP22 transcripts were localized by in situ

hybridization in motor nuclei of certain cranial nerves in the brainstem and in the

motoneurons of spinal nerves localized in the ventral horn of the spinal cord (Parmantier

et al., 1995). During embryonic development PMP22 mRNA was also detected in a

variety of murine tissues (Baechner et al., 1995; Parmantier et al., 1997). Beside the

strong expression found in Schwann cells especially during myelination,

immunoreactivity for PMP22 has been described in dorsal root ganglia (DRG) neurons,

satellite cells, and in the spinal cord predominantly in the dorsal horn (De Leon et al.,

1994).

Recent studies characterized the transcriptional startpoints and methylation pattern in the

PMP22 promoters in tumour cell lines and in the peripheral nerve. They proposed

multiple transcriptional start points for exon 1B and proposed a third promoter in front of

exon 2, which seems to be active in the osteosarcoma and glioblastoma cell lines RH30

and SF763 (Huehne and Rautenstrauss, 2001; Huhne et al., 1999).

For PMP22, so far only initial in vitro studies have been performed on the human 5’

regulatory region, resulting in identification of a cAMP sensitive silencer element


25

between -0.3kb and -3. 5kb relative to the transcription start site of promoter 1 (Saberan-

Djoneidi et al., 2000) in the mouse Schwann cell line MSC80 (Boutry et al., 1992). In

the RT4-D6P2T schwannoma cell line a positive regulatory element (-105 to -43bp) was

described immediately upstream of the minimal promoter 1. This element forms a DNA-

protein complex in vitro (Hai et al., 2001).

3.4 Aim and motivation of the work

To improve our understanding of the mechanisms involved in CMT1A and HNPP and to

determine the molecular and cellular function of the dosage-sensitive gene PMP22, it is

important to determine the role of PMP22 transcriptional regulation during development

and in pathogenesis of the peripheral nervous system. Deciphering the molecular basis of

this regulation is crucial for our understanding of the interplay between neurons and glial

cells in myelin formation in health and in diseases like multiple sclerosis and

neuropathies. Furthermore, such studies may provide the rational basis for potential gene

therapeutic approaches to normalize PMP22 expression in CMT1A (Hai et al., 2001) and

will provide additional insights into the coordinate regulation of myelin genes.

To this end, I started with an additional characterisation of the PMP22 promoter in vitro

(see chapter 4 of the thesis). Second, I performed an extensive and systematic dissection

of the regulatory regions of PMP22 in transgenic mice as presented in chapter 5 and 6 of

the thesis. Finally, the results will be compared with a computer analysis of conserved

sequences and potential binding sites as presented in chapter 7.

MYELIN GENE REGULATION IN VITRO RESULTS PART 1

27

4 MYELIN GENE REGULATION IN VITRO

4.1 Aims and Experimental strategies

The following chapter describes the validation of different cell culture systems to study

PMP22 transcriptional regulation in cell culture. As discussed earlier, a cell culture

system does focus more on the cell intrinsic mechanism of a certain cell type, and

therefore neglects the fact that developing or myelinating Schwann cells integrate a

complex pattern of extracellular signals in addition to their cell intrinsic mechanisms in

the in vivo situation, which then finally results in their coordinate regulation of genes. On

the other hand a well-chosen cell culture system may allow, for example, a rapid

molecular dissection of a regulatory region, the determination of DNA-binding factors,

or the study of certain intracellular pathways by specific activation or inhibition. It is

possible to study certain mechanisms in ways that are either not possible with the

available techniques so far in vivo or are very time consuming. However, the results

obtained in a cell culture system should be confirmed in vivo whenever possible.

Consequently, our aim was to find a system in which some aspects of PMP22 gene

regulation could be studied and modulated in a fast and reliable way.

A classical and frequently used way to characterize and identify negative and positive

regulatory elements in promoter regions is to transiently transfect cell lines with reporter

constructs harbouring different fragments of the regulatory region of the gene. For

PMP22, initial experiments, performed with constructs specific for promoter 1 or

promoter 2 driven expression of PMP22, already indicated that promoter 2 of PMP22

shows stronger transcriptional activity in cultured Schwann cells than does the myelin

associated promoter 1 (Suter et al., 1994). To gain more insight into PMP22

transcriptional regulation in cultured Schwann cells, which correspond, based on their

molecular markers, to non-myelinating Schwann cells, we performed a more extensive

promoter deletion study, similar in design to previous studies of other myelin gene

promoters such as the Myelin basic protein (MBP) (Li et al., 1994; Miura et al., 1989;

Tamura et al., 1989), myelin protein zero (MPZ) (Brown and Lemke, 1997; Lemke et al.,

1988) or proteolipid protein (PLP) promoters (Berndt et al., 1992; Nave and Lemke,


28

1991; compare also Tab. 1). The results obtained from such studies depend – among

other things – on the cell type in which the transfection studies are performed. In

principle two different strategies can be followed to select a cell type. Either one uses a

non-specific cell line where only basal promoter activity can be expected for the gene of

interest, and subsequently tries to modulate this basal activity by the cotransfection of

additional regulatory factors. The second approach is to choose a cell line which

resembles the cell type in which the gene of interest is as highly expressed as possible.

With this choice one anticipates that additional endogenous factors from the cell line

contribute to the complex transcriptional regulation of the gene. We decided to use the

second approach, using the mouse Schwann cell line MSC80 (Boutry et al., 1992) which

does express PMP22. As a second cell line we selected NIH3T3 fibroblasts, since

previous studies characterized PMP22 regulation in this cell type (Fabbretti et al., 1995;

Manfioletti et al., 1990; Zoidl et al., 1997). The results of this study are presented below

in section 4.2.

In the case of MPZ a successful identification of regulatory factors was possible in cell

culture either in cotransfection assays (Monuki et al., 1990; Zorick et al., 1999) or with

doxycycline-inducible expression of transcription factors in N2A cells (Peirano et al.,

2000). The latter study revealed a clear induction of MPZ expression specifically by

Sox10. We wondered whether we could find coregulation of PMP22 in this system. The

Sox10 induction can be performed in cell culture media with very low level of fetal calf

serum (FCS), an agent known from earlier studies to prevent upregulation of PMP22

when present at high concentrations (Suter et al., 1994). Similarly PMP22/gas3 is

upregulated upon growth arrest of the cells which can be induced by serum withdrawal

(Manfioletti et al., 1990; Schneider et al., 1988). In a first attempt we tried classical

cotransfection experiments with Krox20, Sox10, pax3, Oct-6/SCIP, Ski, and

combinations of these, together with a PM22-lacZ reporter construct (-10/0kb PMP22

lacZ, see Fig. 4-1). In the presence of 10% FCS, reporter gene expression was either not

influenced or inhibited by these factors. This corresponds to observations that others

have made in cell culture (e.g. Bharucha et al., 1994; Monuki et al., 1993). In the

absence of serum we faced the problem that Schwann cells are not efficiently

transfectable, as they stop proliferating after serum withdrawal. Therefore this system


29

was not reliable, and the positive results obtained in certain experiments were not

reproducible (data not shown).

As a next system we analysed the regulation of PMP22 in N2A cells stably transfected

with a construct allowing induction of Sox 10 expression by doxycycline, as established

by Peirano and coworkers (2000). The results obtained in collaboration with this group

are presented in part 4.3 below.

While the in vivo experiments presented in chapters 5 and 6 were in progress, Nagarajan

and coworkers found a way to circumvent the low transfection efficiency in non-

proliferating Schwann cells in the absence of serum. They used an adenoviral system to

overexpress Krox20 in rat and mouse Schwann cells and could indeed observe

upregulation of different myelin genes including PMP22. Nagarajan et al., also presented

a list of Krox20-regulated genes identified in a whole genome expression study

(Nagarajan et al., 2001). We established this in vitro system in which PMP22 can be

upregulated directly by a transcription factor, and could reproduce the upregulation of

myelin genes by Krox20. In addition we confirmed our assumption that upregulation of

PMP22 is reduced in the presence of serum. With this system, we started to test Krox20-

regulated transcription factors, based in part on the published gene chip data (Nagarajan

et al., 2001) including other candidates from other gene profiling experiments (Araki et

al., 2001). The results, which still have preliminary character but for some candidates are

very promising for future studies, are presented in section 4.4 below.

4.2 Promoter Deletion study in MSC80 and NIH 3T3 cells

As a first step, we generated the basic PMP22 promoter construct (-10/0kb PMP22 lacZ)

using the sh ble-lacZ fusion gene driven by promoter 1, promoter 2 and upstream

sequences of the mouse PMP22 gene. Ten kb of 5’-flanking PMP22 DNA were used,

comprising 5.8 kb upstream of exon 1A including promoter 1, 2.3 kb between exon1A

and exon 1B including promoter 2, and 1.5 kb downstream of exon 1B with the first 43

bp of exon 2 (see Fig. 4-1). In a next step we generated additional reporter constructs

harbouring a 5’ deletion of the first 6kb region (-4/0 kb Pro2 lacZ), or the first 7kb region

(-3/0 kb Pro2 lacZ) leading to a reporter construct containing only promoter 2 with 1kb


30

or 2kb sequences upstream of exon 1B, respectively (see Fig. 6-1). The deletion of exon

1A or the exon 1B with the associated promoter resulted in the constructs Del1AlacZ and

Del1BlacZ (see Fig. 4-1). In the Del1AlacZ construct we deleted the -230 bp to +330 bp

and in the Del1BlacZ construct the -170 bp to +410 bp region in relation to the

transscription start site as mapped by Suter, et al. (1994). All these construct were

transfected in the mouse Schwann cell line MSC80 (Boutry et al., 1992) and in NIH 3T3

cells. We used equimolar amount of each reporter construct and corrected for different

transfection efficiencies by cotransfection of a constant amount of SV40 luciferase

plasmid. The measured ß-gal activities are shown in relation to the SV40-lacZ expression

(set to a 100% in both cell lines; Fig. 4-1).

The mouse Schwann cell line (MSC80) is one of the few Schwann cell lines derived

from mice. It was established from purified mouse Schwann cell cultures using high

levels of fetal calf serum. Most of the MSC80 cells are bipolar or stellate (3-5 processes)

in shape, and express antigens of Schwann cells such as S-100 and laminin but also the

non-myelinating Schwann cell antigen GFAP. In vivo after transplantation in or at a

distance from a lysolecithin-induced lesion, MSC80 cells form myelin around the

demyelinated host axons (Boutry et al., 1992).

As a general finding, higher expression levels were seen in the MSC80 cell line

compared to the NIH3T3 fibroblasts. Highest relative reporter gene activity, with levels

about two fold above the SV40 promoter, can be found with the -10/0kb PMP22 lacZ

construct in MSC80 cells in the presence of forskolin. Forskolin increases the

intracellular levels of cAMP and is thought to mimic some aspects of axonal contact of

Schwann cells. In our experiment, forskolin increased PMP22 promoted reporter gene

expression about twofold in MSC80 cells, whereas in NIH 3T3 cells no consistent effect

of forskolin was observed.

The observation that deletion of promoter 2 and exon1B (construct Del1BlacZ, Fig. 4-1)

led to a complete loss of expression and that the deletion of exon 1A and its associated

promoter 1 (construct Del1AlacZ, Fig. 4-1) did not change expression levels

substantially, shows that mainly promoter 2 is active in the MSC80 cell line. The same

was observed in NIH 3T3 cells. Indeed, quantitative RT-PCR with RNA from MSC80

cells transfected with the -10/0kb PMP22 lacZ construct, using primers specific for

exons 1A or 1B and either the reporter or the endogenous gene, confirmed that promoter


31

.

Figure 4-1: Activity of different PMP22-promoted reporter constructs in the mouseSchwann cell line MSC80 (grey bars) and in NIH3T3 fibroblast cells (black bars)(a), and mRNA expression levels of transfected lacZ constructs and endogenousPMP22 (b). (a) The relative expression of the lacZ reporter gene obtained 40 hoursafter transfection is shown for each construct in the presence (+) or absence (-) of20µM forskolin. The expression levels of the SV40 promoter in the absence offorskolin were defined as 100%. (b) mRNA expression levels of 1A-lacZ and 1B-lacZ transcripts derived from the transfected -10/0kb PMP22 lacZ construct andthe 1A-PMP22 and 1B-PMP22 transcripts of the endogenous PMP22 promoters inMSC80 cells, determined by quantitative RT-PCR (see Fig. 5-1). The cells weremaintained in DMEM containing 10%FCS. Error bars represent the SD of valuesobtained from three independent transfections.

10

0%

Promoter activity%SV40-LacZ (ß-gal activity/Luminescence)

1A 1B 2lacZ

-10/0 kb lacZ

1A 1B 2-4/0 kb Pro2 lacZ lacZ

1B-3/0 kb Pro2 lacZ 2lacZ

neg. control (pUT111) lacZ

1B 2Del1A lacZ lacZ

-170 +410

1A 2Del1B lacZ lacZ

-230 +330

SV40 lacZ lacZSV40F

orsko

lin 2

0µ

M

-

+

-

+

-

+

-

+

-

+

+

-

+

-

MSC80 cells

NIH 3T3 cells

0 50 100 150 200 250

10-6

10-5

10-4

10-3

10-2

10-1

log

[R

NA

]/[G

AP

DH

]

1A-la

cZ

1B-la

cZ

1A-P

MP

1B-P

MP

a

b

-

+

-

+

-

+

-

+

-

+

+

-

+

-


32

2 is about 100 times more active than promoter 1, both for the transfected and the

endogenous gene (Fig. 4-1b, for details of the quantitative RT-PCR see Fig. 5-1 below).

The differential promoter activities also did not change considerably in MSC80 cells

upon addition of 10µM forskolin to the cell culture medium (data not shown). This is in

contrast to the situation in the sciatic nerve, where about four times more 1A-PMP22

than 1B-PMP22 mRNA was detected (compare Fig. 5-6). This supports the hypothesis

that promoter 1 is the myelin-associated promoter and therefore only basal levels of 1A-

PMP22 mRNA are detected in non-myelinating Schwann cells. In other words: the

myelin specific promoter 1 is active only at very basal levels in cultured Schwann cells.

4.3 Is PMP22 regulated by Sox10 in N2a cells?

Peirano et al., (2000) showed that MPZ can be upregulated by Sox10 in N2A

Neuroblastoma cells. We therefore asked whether PMP22 might be similarly regulated.

In collaboration with the group of M. Wegner (Erlangen), N2A neuroblastoma cells

expressing the reverse tetracycline-controlled transactivator (rtTA) and the cDNA of

Sox10 under control of a tetracycline-regulatable promoter were induced with

doxycycline or vehicle alone in DMEM Medium containing 0.5% FCS for 48 hours

before harvesting the RNA. A high induction of MPZ upon doxycycline induced

expression of Sox10 (+ Dox) compared to non-induced control (- Dox) can be observed

(Fig. 4-2a,b; Peirano et al., 2000; for technical review see Mansuy and Suter, 2000). In

contrast to MPZ neither a substantial upregulation of the very low levels of the 1A-

PMP22 mRNA derived from promoter 1 (Fig. 4-2b), nor of the 1B-PMP22 mRNA

derived from promoter 2 (Fig. 4-2c) can be detected by quantitative RT-PCR. As a

control, the total amount of PMP22 mRNA was determined with a different set of

TaqMan Primers specific for the translated region and detecting all PMP22 mRNA

species. Also in this case total levels of PMP22 mRNA did not show any upregulation, as

expected since the total PMP22 mRNA consists again mainly of 1B-PMP22 transcripts.

In fact, the opposite was the case: a weak inhibition of PMP22 expression was observed


33

(Fig. 4-2e). As a conclusion, PMP22 seems not to be co-regulated with the myelin

protein zero under these conditions.

Figure 4-2: mRNA expression levels of Sox10 (a), Myelin Protein Zero (b, MPZ)and PMP22 (c, d, e) with (+Dox) or without (-Dox) doxycycline induction of Sox10in N2a cells, detected with semiquantitative (a,b) or quantitative (c-e) RT-PCR.

4.4 Screening for transcription factors upregulated uponKrox20 induced myelin gene expression

Schwann cells are well infectable with relatively low titres of adenovirus. In contrast to

retrovirus, adenovirus can also infect non-proliferating cells (Fig. 4-3a-c). Therefore

infection with adenoviral vectors offers a very efficient and reliable way to express, for

example, a transcription factor in nonproliferating Schwann cells. Nagarajan and

coworkers (2001) exploited this fact and studied changes in gene expression after

infection of rat Schwann cell with an adenovirus expressing Krox20 and eGFP under

control of a CMV promoter (Ehrengruber et al., 2000). They used microarrays in a

global analysis of gene expression in rat Schwann cells infected with an adenovirus

expressing Krox20, as compared to control-virus infected Schwann cells (Nagarajan et

al., 2001). Our hypothesis was that transcription factors regulated by Krox20 are directly

or indirectly involved in the regulatory network of myelin gene regulation. Based on the

published gene chip data, educated guesses and other gene profiling studies (Araki et al.,

2001), we started to confirm the Krox-20 regulation of our candidates. These included

the KRAB-zinc finger protein KZF-1 and KZF-1 like ((Bellefroid et al., 1998), accession

0

0.06

0.12

0

0.08

0.16

0

0.0001

0.00025

1B-P

MP

22/G

AP

DH

PM

P22

tot /

GA

PD

H

Dox +

a b c d

-+- +-+- +- +- +-

1A-P

MP

22/G

AP

DH

Sox10 MPZe


34

Figure 4-3: Quantitative RT-PCR screening for Krox-20 regulated genes in ratSchwann cells upon infection with an adenovirus expressing Krox-20. (a-c) Highexpression levels of GFP encoded by the viral vector (taken as a surrogate formeasuring Krox20 directly) can be detected in nearly all rat Schwann cells infectedin DMEM medium in the presence of 10% FCS with a dilution of 1:100 (a) or 1:10(b) of the Adegr2GFP virus (Ehrengruber et al., 2000) after 48 hours. When theinfection was performed in defined N2 medium (c) with the same dilution of thevirus (1:10), weaker expression levels were observed, which nevertheless inducedthe upregulation of PMP22 and periaxin mRNA (d). (d) Differences in mRNAexpression levels (fold difference) of Krox20 versus control-virus infected Schwanncells, 30 hours (experiment 1), 24 or 48 hours (experiment 2) after infection, assayedwith triplicate measurement of the indicated TaqMan or SYBRGreen PCR(corresponding dissociation curve shown in the last column). Abbreviation: n.a.(ME): not applicated; dissociation curve done by M. Ehrengruber.


35

number AF175222 and BC004747 for the mouse cDNA), the broadly expressed POU

domain transcription factor Brn-2 (Donahue and Reinhart, 1998; Schreiber et al., 1997),

the Iron responsive element binding protein (IREBP; accession number X61147), Egr1

and Egr3 to confirm the effect of specific induction of Egr2/Krox20 and the Schwann

cell marker S100 as a control which should show unchanged levels (Fig. 4-3d).

In the absence of serum and within 48 hours after infection, we could confirm the

upregulation of total PMP22 levels, and an upregulation of the 1B-PMP22 message. The

1A-PMP22 levels were strongly increased (20-80 fold), but it must be emphasized that

even the induced levels were still low. We did not observe a robust induction of PMP22

in the presence of serum. At least in experiment 2 we detected a seven-fold induction of

Periaxin expression (Fig. 4-3d).

In regard to the screening we could only confirm the two-fold up-regulation of Brn-2, but

not of KZF-1, which in the published gene chip analysis was upregulated about five-fold

(Nagarajan et al., 2001). No regulation was observed for IREBP. The negative first

derivative of the fluorescence vesus temperature curve of the PCR products (Fig. 4-3d)

shows only a single peak for IREBP, indicating that only one single product was

amplified. In the case of KZF-1 a more sensitive PCR might show different results since,

the gene was expressed at very low levels in cultured Schwann cells. In contrast to the

gene chip data where a five-fold upregulation of Egr1 was detected, we determined in our

experiment either unchanged or decreased levels of Egr1 upon Krox20 expression.

Expression levels of egr3 seem not to be influenced by Krox20 expression.

In principle, the system can be used to study the regulatory network of PMP22 gene

regulation in cell culture, since we are able to induce myelin gene expression in this in

vitro system. In addition, in the case of Brn-2 our screening approach revealed a valuable

candidate for further studies, as Brn-2 regulation in mice was confirmed in the meantime

by (Sim et al., 2002).

-10/0KB PMP22 LACZ CONSTRUCT IN VIVO RESULTS PART II

37

5 REGULATORY ELEMENTS ON THE -10/0KB PMP22LACZ TRANSGENE

5.1 Aims and Experimental strategies

Cultured Schwann cells without coculturing with neurons do not myelinate in vitro, so

that the Schwann cells do not integrate axonal signals into their transcriptional

regulation. Therefore a systematic dissection of the regulatory regions in transgenic mice

provides an excellent assay to study gene regulation involving numerous developmental

and physiological signals. Consequently, we examine the role of the PMP22 promoters

and their 5’ regulatory sequences on the transcriptional regulation of PMP22 in vivo

which is described in the following section of my thesis. To this end, we produced

transgenic mice in which promoters 1 and 2 drive expression of a lacZ reporter gene,

either in wildtype mice or in animal models for CMT1A.

Part of the following results have been published in:

Maier, M., Berger, P., Nave, K. A., and Suter, U. (2002). Identification of the Regulatory

Region of the Peripheral Myelin Protein 22 (PMP22) Gene That Directs Temporal and

Spatial Expression in Development and Regeneration of Peripheral Nerves.

Molecular and Cellular Neuroscience (MCN) 20: 93-109.


38

5.2 Generation of transgenic mice expressing the reporterlacZ gene under the control of the PMP22 genepromoters (-10/0 kb PMP22 LacZ)

Transgenic mouse lines were generated using the sh ble-lacZ fusion gene driven by

promoter 1, promoter 2 and regulatory sequences of the mouse PMP22 gene (Suter et al.,

1994). Ten kb of 5’-flanking PMP22 DNA were used including promoter 1 with exon1A,

promoter 2 with the corresponding exon 1B and 1.5 kb downstream of exon 1B with the

first 43 bp of exon 2 (Fig. 5-1) (Suter et al., 1994; van de Wetering et al., 1999). The sh

ble gene confers resistance to the antibiotics of the zeomycin group (Drocourt et al.,

Figure 5-1: Generation of PMP22 Promoter-sh ble-lacZ (-10/0kb PMP22 lacZ)transgenic mice. Diagram of the endogenous mPMP22 genomic locus and the -10/0kb PMP22 lacZ transgene. (a) Promoter 1 is located in front of the non-translatedexon 1A, Promoter 2 in front of the alternatively used exon 1B. All numbers refer tothe nucleotide +1 which was defined at the translation start codon on exon 2 (ATG).The specific forward and backward primers for the detection of the twoendogenous mRNAs, 1A-PMP22 (Pr1A and Pr2PMP22) and 1B-PMP22 areindicated (Pr1B and Pr2PMP22). (b) To generate a reporter transgene under thecontrol of the PMP22 promoters, the endogenous PMP22 sequence was replaced atthe translation start site with the coding sequence of a sh ble-lacZ fusion reportergene. The specific primers for the two transgenic mRNA species 1A-lacZ (Pr1A,Pr2lacZ), 1B-lacZ (Pr1B, Pr2lacZ) are indicated. The PMP22 TaqMan probe (blackrectangle with 5’ reporter and 3’ quencher dye modifications) used forquantification is located in the first half of exon 2 in a sequence common to all fourdifferent RNA transcripts.

1A 1B 2 3

1A-PMP221B-PMP22

1A-lacZ1B-lacZ

Pr1APr1B

Pr2PMP22

4 5ATG

Promoter1 Promoter2

1 kb

a

b

1A 1B

Pr1APr1B

Pr2lacZ

Promoter1 Promoter2 sh ble lacZ pA

ATG

-10kb


39

1990; Gatignol et al., 1988) and the addition of the Escherichia coli lacZ gene in frame

to the sh ble sequence allows monitoring of PMP22 gene expression. DNA extracted

from the tails of 85 offspring mice was analyzed by PCR for the presence of the

transgene which was detected in 25 animals. Expression of the transgene in innervating

peripheral nerves of the tail was analyzed on cross-sections derived from biopsies of the

founder animals. Three founders showed high lacZ expression levels and seventeen

showed a moderate or weak expression. Five founders were mated for further analysis

with B6D2F1 hybrid mice and stable lines with high (lines 48.4, 44.2, 49.3), moderate

(line 37.1) or low (line 45.2) expression levels were established. The expression levels

were confirmed on whole mount X-gal stainings of sciatic nerve in F1 animals (data not

shown) which were then bred again with hybrid mice. Lines 48.4 and 44.2 were analyzed

in detail and showed identical patterns of expression of the transgene both in the embryo

and adult. With all five lines, transgenic animals were fertile and appeared normal

throughout live.

5.3 Analysis of the first wave of -10/0kb PMP22 lacZexpression during embryonic development

ß-gal expressing cells were first detected at E10.5 in the limb buds of the developing

fore- (Fig. 5-2a, arrowhead) and hindlimbs as well as at the midbrain-hindbrain border

(Fig. 5-2a, arrow). The expression pattern at E11.5 (data not shown) was similar to that

seen at E12.5 (Fig. 5-2b). At E12.5 a prominent staining was detected in the lateral and

medial regions of the limbs (Fig. 5-2b, arrowhead). This expression of the -10/0kb

PMP22 lacZ transgene is consistent with the finding of endogenous PMP22 RNA by in

situ hybridization in the developing limbs (Baechner et al., 1995). Furthermore, we

detected ß-gal expression in the ventricular epithelium of the rhombencephalon, but

neither more rostrally in the myencephalon or prosencephalon, nor more caudally in the

alar plate of the spinal cord, either at E12.5 (Fig. 5-2b) or at E14.5 (Fig. 5-2c, d, f). These

results are in agreement with the endogenous expression pattern of the PMP22 gene in

the developing CNS as determined by (Parmantier et al., 1997). This is also true for some


40

weak X-gal staining that we detected in the neuroepithelium of the olfactory bulb and in

some tongue muscles (data not shown). The spatial expression pattern at E14.5 (Fig. 5-

2d,e) is similar to the pattern at E12.5. On whole-mount X-Gal stainings (Fig. 5-2c) or

sections of E14.5 embryos, we detected additional prominent staining in the outer ear

and scattered blue cells were found in the skin and in the retina of the eye (data not

shown). This finding contrasts to the data of (Baechner et al., 1995) who localized

endogenous PMP22 mRNA predominantly to the inner ear and the lens of the eye. How-

Figure 5-2: ß-gal expression during embryonic development of -10/0kb PMP22 lacZtransgenic mice. Whole mount X-gal stainings from embryonic day E10.5 (a), E12.5(b) and E14.5 (c). Histological X-Gal stainings of longitudinal (d) and transverse (e,f) cryosections at E14.5 at the levels of the embryo as indicated in (c). At E10.5 (a) ß-gal activity was detected in the outer and inner region of the limbs (arrowhead) andin the region of the midbrain-hindbrain boundary (arrow). At E12.5 (b) and E14.5(c, d, e, f) ß-gal activity could be detected most prominently in the limb muscles(arrowhead in b, d, e), in the roof of the midbrain (f), in the external ear (arrow inc) and in some scattered cells of the olfactory epithelia and epidermis. mv:mesencephalic vesicle. Scale bars: 500µm (a-e), 100µm (f).

ba cf

e

mv

fed mv


41

ever, the pattern of transgene expression was consistent in our transgenic lines (data not

shown). Thus, it is likely that negative regulatory elements that inhibit the expression of

the endogenous PMP22 gene in these areas are missing on our transgene.

In accordance with PMP22 mRNA in situ studies (Baechner et al., 1995; Parmantier et

al., 1997) a reduced expression of the transgene was observed in late embryonic

development (data not shown). At E16.5 and E18.5 remaining ß-gal expression could be

detected in the outer ear and in the limb muscles of -10/0kb PMP22 lacZ transgenic

animals.

5.4 -10/0kb PMP22 lacZ expression is strongly upregulatedin postnatal neurons and Schwann cells of peripheralnerves

The second wave of -10/0kb PMP22 lacZ transgene expression starts around birth. First

ß-gal positive cells in the DRG can be detected at E19.5 on whole mount stainings (data

not shown). An increasing number of ß-gal positive DRG neurons can be observed on

sections at postnatal day 1 (P1, Fig. 5-3a) and P5 (Fig. 5-3b, inset). At this age, X-gal

staining was also observed in the cartilage regions of the bones (Fig. 5-3b, asterisk), as

had been observed for PMP22 mRNA during late embryogenesis (Baechner et al., 1995).

Transgene expression in the muscles persisted throughout postnatal development, but in

contrast to the embryonic expression at E14.5, staining was detectable additionally in

muscles outside of the limbs and was more localized around single muscle fibers (Figs.

5-3a, b, arrowhead).

Postnatal day 10 (P10) was the first time at which ß-gal expressing cells were seen in the

ventral horn of the spinal cord (data not shown). At P21, the ß-gal activity was more

intense than at P10, and on sections through the thoracic and lumbar spinal cord (Fig. 5-

3c, d) a diffuse X-gal staining was additionally observed in the dorsal horn of the spinal

cord, mainly located in the area of lamina II. Also the dorsal column tracts, especially the

gracile fasciculus as well as some large cell bodies in the ventral horn of the gray matter,

were positive for ß-gal. These large, multipolar cells have been described to express


42

endogenous PMP22 mRNA and were identified as motor neurons by (Parmantier et al.,

1995).

Figure 5-3: Developmental appearance of ß-gal-positive cells in DRGs and spinalcords of -10/0kb PMP22 lacZ transgenic animals. X-Gal stainings of spinal cordcross-sections of P1 (a) and P5 (b) mice. Expression of lacZ starts around birth andis upregulated during the first postnatal week in the DRG (arrows in a, inset in b).During the first postnatal days, additional expression can be detected aroundmuscle fibers (arrowhead in a, b) and in the cartilage regions of the bones (asteriskin b). (c, d) At P21, lacZ expression is seen in the ventral horns of the grey matter insome motor neurons (mn), in the sensory neurons of the lamina II, in the dorsalcolumn neurons and in the ventral roots at thoracic level of the spinal cord (c) aswell as at lumbar level (d). (e) Immunohistochemical stainings of dissociated DRGsfrom E19 embryos. Neuro-filament (NF, red) and ß-galactosidase (ß-gal, green)were detected in the same cells. (f) X-Gal histochemistry on teased nerve fibers of aten week-old -10/0kb PMP22 lacZ transgenic mouse. Virtually all fibers with large(1, inset), medium (2, inset) and small (3, inset) caliber axons were enwrapped witha ß-gal positive Schwann cell, albeit with different level of expression. mn:motoneurons, vr: ventral roots, sn: sensory neurons, gt: gracile tract. Scale bar:200µm (a-d), 20µm (f)

1

2 3

��

��

b

c

��

��

a

��

d

NF ß-gal

e

��

*

f


43

DRG neurons give rise to a long peripheral axon and a shorter central axon. Depending

on the sensory input, the central axons terminate either directly in the dorsolateral region

of the spinal cord, or ascend ipsilaterally through the dorsal columns of the cord and

terminate in the dorsal column nuclei located in the lower medulla. Thus, it appears

likely that the dot-like X-gal staining found in lamina II and the staining in the dorsal

column tracts derive from ß-gal that has diffused into the projections of the sensory DRG

neurons since the cell soma of these neurons are strongly ß-gal positive. This

interpretation is supported by the analysis of longitudinally cut axons at places where

they are myelinated by ß-gal-negative oligodendrocytes (Fig. 5-4d, asterisk), for example

at the central-peripheral nervous system transition zone of the trigeminal nerve.

Additional evidence that the ß-gal positive DRG cells are neurons was provided by

analysis of the ß-gal expression of the different cell types in vitro. DRG from E19.5

embryos were dissected, enzymatically dissociated, and plated on collagen-coated cell

culture dishes. Staining of the mixed cell population after three days with antibodies

against ß-gal and neurofilament revealed localization of ß-gal immunoreactivity in DRG

neurons (Fig. 5-3e). All neurofilament positive cells on the cell culture plate also showed

immunoreactivity for ß-gal. In addition, X-gal staining of sister plates showed ß-gal

positive cells identified by morphology as neurons with variable staining intensities (data

not shown).

Transgene expression in Schwann cells of the sciatic nerve was first detected around P8

with a few scattered blue cells visible in whole mount preparations. During peak period

of myelination around P10, the number of stained Schwann cells as well as the staining

intensity increased dramatically (data not shown). At P21, ß-gal positive Schwann cells

were found in the spinal nerves (for sciatic nerve, see Fig. 5-3f) as well as in some

cranial nerves (for trigeminal nerve, see Fig. 5-4d). The staining intensity of the

Schwann cells was considerably stronger than that of the motor and DRG neurons, and

was maintained in old adult animals (up to 1 year of age). ß-gal positive Schwann cells

showed comparable staining intensity in the dorsal and ventral roots of the spinal cord

(Fig. 5-3c, only ventral roots visible on this section).


44

X-Gal stainings on teased nerve fibers of adult -10/0kb PMP22 lacZ transgenic animals

showed expression of the reporter gene in the majority of the fibers (Fig. 5-3f). Virtually

all Schwann cells are positive for X-Gal whether they enwrap relatively small (inset Fig.

5-3f, fiber No. 3), medium (inset Fig. 5-3f, fiber No. 2) or large caliber axons (inset Fig.

5-3f, fiber No. 1). Accumulation of cytoplasmatic ß-gal can be observed in regions with

increased Schwann cell cytoplasm, the paranodes (arrow in inset, Fig. 5-3f) and

Schmidt-Lanterman incisures (arrowheads in inset, Fig. 5-3f).

To confirm the onset of transgene expression and to quantitate the expression of the -10/

0kb PMP22 lacZ transgene, the ß-gal enzymatic activity was determined in homogenates

of the sciatic nerve at different time points (Fig. 5-5a). Strong upregulation of the

transgene was observed after P10. This confirms the rather late upregulation of the

transgene seen in X-gal stainings compared to what has been described for the

endogenous PMP22 mRNA in rat sciatic nerve (Suter et al., 1994). To compare the

transcriptional regulation of the endogenous PMP22 with the transgene in the sciatic

nerve of mice, transgene expression was determined directly on the mRNA level. For this

purpose, the levels of the exon 1A-containing-lacZ (1A-lacZ) mRNA and the exon 1B-

containing-lacZ (1B-lacZ) mRNA was determined by quantitative RT-PCR, and

compared with the expression of the exon 1A-containing-PMP22 (1A-PMP22) mRNA

and the exon 1B-containing-PMP22 (1B-PMP22) mRNA (Fig. 5-1). GAPDH mRNA

was used as an internal standard. The lacZ mRNA species in transgenic animals (Fig. 5-

5b) were compared to PMP22 mRNA from wildtype littermates (Fig. 5-5c) at different

time point during postnatal development. The quantitative RT-PCR for the lacZ mRNA

confirmed the delayed upregulation of the -10/0kb PMP22 lacZ transgene around P10

also at the transcriptional level. The expression profile of the endogenous PMP22 mRNA

in wildtype mice was similar to that found in rats during myelination (Suter et al., 1994)

with the main upregulation of endogenous PMP22 mRNA occuring between postnatal

days P1 and P10 during the onset of myelination (Fig. 5-5c).

We also compared the relative contributions of the endogenous promoters 1 and 2 to the

regulation by the exogenous promoters 1 and 2 of the transgene (in two different lines).

No significant difference in the ratio of 1A to 1B transcripts was seen for the lacZ-


45

containing transcripts compared to wildtype PMP22 mRNA (Fig. 5-6b, c). Furthermore,

the amount of endogenous PMP22 mRNA was similar in both wildtype and transgenic

mice, indicating that the expression of the reporter gene does not alter the endogenous

PMP22 mRNA levels (data not shown). This is consistent with the fact that no phenotype

was observed in the transgenic animals and indicates that the presence of additional

copies of the PMP22 promoter region in the genome does not have a major influence on

the transcriptional regulation of the endogenous PMP22 gene. The absolute abundance

of endogenous PMP22 and transgenic lacZ mRNA can therefore be compared directly

revealing that, at P21, the lacZ mRNA is approximately two orders of magnitude less

abundant than the PMP22 mRNA (Fig. 5-5b, c).

5.5 Transgenic ß-gal expression in sensory and motorneurons of cranial nerves

Serial transverse cryosections through the brainstem of three week-old -10/0kb PMP22

lacZ transgenic mice revealed that ß-gal positive cells can also be detected in several

cranial nerve nuclei of the brainstem (Fig. 5-4a, b, c). Except for the nuclei of the

occulomotor (nerve III), trochlear (nerve IV), abducens (nerve VI) and auditory nerve

(nerve VIII), where no signal was detected, all other motor and sensory nuclei of cranial

nerves showed a definitive X-gal staining. This is illustrated for the hypoglossal and

dorsal vagus nuclei (nerve XII and X, respectively, Fig. 5-4a), the motor nuclei of the

facial (VII) nerve (Fig. 5-4b), and the motor nuclei of the trigeminal (V) nerve (Fig. 5-

4c). This expression pattern of the transgene is in accordance with the endogenous

expression of PMP22 (Parmantier et al., 1995) which was found in the same subset of

nuclei of cranial nerves.

In Figure 5-4d, the different staining intensities of neuronal and Schwann cell expression

of the -10/0kb PMP22 lacZ reporter gene are illustrated for the trigeminal nerve at the

border between the central and peripheral nervous system (arrowhead in Fig. 5-4d). As

soon as the cranial nerve neurons are myelinated by Schwann cells the staining intensity


46

increased dramatically, and confirms that oligodendrocytes around these ß-gal positive

neurons do not express the transgene in detectable amounts.

Figure 5-4: ß-gal expression in spinal nerve nuclei in the brainstem of -10/0kbPMP22 lacZ transgenic mice. At P21, X-Gal staining is found in the hypoglossal (a,arrows) and dorsal vagus nuclei (a, arrowhead), in the motor nuclei of the facialnerve (b, arrows) and in the trigeminal nerve nuclei of the brainstem (c, arrows). Inaddition, expression of the transgene was detected in the sensory fibers of the spinaltrigeminal tract (a, asterisks) and in the region of the mesencephalic nuclei of thetrigeminal nerve (c, arrowheads). (d) The neuronal expression (d, asterisk) is muchweaker than the expression by myelinating Schwann cells as shown on longitudinalsections of the trigeminal nerve at the transition zone between the central andperipheral nervous system (arrowhead). Scale bar: 200µm

b

d

*c

a*


47

5.6 Tissue specificity of the -10/0kb PMP22 lacZ transgene

In a next step, we investigated whether transgene expression is also found in non-neural

organs. In earlier studies, endogenous expression of PMP22 mRNA outside the nervous

system was detected in small intestine, lung, to some extent in heart by RNase protection

assays (Suter et al., 1994) and in a variety of other tissues by RT-PCR (Parmantier et al.,

1995; van de Wetering et al., 1999). Since no specific staining was observed on sections

of heart, lung, intestine and liver (data not shown), we decided to use the sensitive

luminescence ß-gal assay. We determined the ß-gal activity in homogenates of brainstem,

lung, intestine, muscle, heart, liver and sciatic nerve in -10/0kb PMP22 lacZ transgenic

animals compared to background levels in wildtype animals (Fig. 5-6a). Beside the

strong promoter activity in the sciatic nerve, we detected ß-gal activity significantly over

background of non-transgenic littermates only in brainstem and muscle tissue, although

the expression levels were about four orders of magnitude lower than in the sciatic nerve.

ß-gal activity in lung, intestine and heart was not distinguishable from background and

was not detectable at all in liver.

To confirm the expression of endogenous PMP22 mRNA in these organs, quantitative

RT-PCR was performed. High expression of 1A- and 1B-PMP22 mRNA in the sciatic

nerve and lower expression levels of 1B-PMP22 mRNA in lung and intestine (Fig. 5-6b)

were observed, in accordance with data from the rat system (Suter et al., 1994). In

contrast, no lacZ mRNA was detectable in lung or intestine. Whether this is due to the

general low abundance of the lacZ mRNA at levels below the limit of reliable detection

with RT-PCR, or due to missing promoter activity remains open.


48

Fig. 5-5 Fig. 5-6

Figure 5-5: Temporal expression profile of ß-galactosidase in the sciatic nerveduring postnatal development. (a) ß-gal activity in homogenates from the sciaticnerve increases dramatically in parallel with myelination. Four to six sciatic nerveswere analyzed. (b) Total RNA was isolated from sciatic nerve of -10/0kb PMP22lacZ transgenic animals at postnatal days 0, 4, 8, 10, 21 and 60, and the amount oftransgenic 1A-lacZ or 1B-lacZ mRNA per GAPDH mRNA was determined byquantitative RT-PCR. Total mRNA levels correlate well with the regulation of lacZexpression. Error bars represent the SD of six values obtained from twoindependent experiments. ß-gal activities are shown in relative light unit(RLU)*106. (c) Endogenous levels of 1A-PMP22 and 1B-PMP22 transcripts asmeasured by quantitative RT-PCR.

Figure 5-6: Spatial distribution of PMP22 promoter activity. (a) ß-gal expressionlevels in organs of three week-old -10/0kb PMP22 lacZ transgenic animals andwildtype littermates. The highest expression levels by far were found inhomogenates of the sciatic nerve. About four orders of magnitude less but stillsignificant levels of activity could be detected in brainstem and muscle tissue. Errorbars represent the SD of the ß-gal activities/µg protein of three different animals.

0

2

4

6

8

0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 100

wildtype PMP22-lacZ

ß-g

al a

ctiv

ity/�

g p

rote

ina

days postnatal

0

4

8

1 2

0 1 0 2 0 3 0 4 0 5 0 6 0 7 0

1A-PMP221B-PMP22Total (1A+1B)

[PM

P22

mR

NA

] / [

GA

PDH

mR

NA

]

b

[lac

Z m

RN

A]

/ [G

APD

H m

RN

A]

0

0.01

0.02

0.03

0 1 0 2 0 3 0 4 0 5 0 6 0 7 0

days postnatal

1A-lacZ1B-lacZTotal (1A+1B)

days postnatal

c

a

b

0

4

8

1 21A-PMP22 1B-PMP22

[PM

P22

mR

NA

] /

[GA

PDH

mR

NA

][l

ac Z

mR

NA

] /

[GA

PDH

mR

NA

]

0

0.01

0.02

0.03

sciatic nerve

liver lung intestine

1A-lacZ Line 44.2

1B-lacZ Line 44.2

1A-lacZ Line 48.4

1B-lacZ Line 48.4

sciatic nerve

liver lung intestine

0

2

4

6

8

40000

sciaticnerve

brain-stem liver lung intes-

tine muscle heart

wildtype

ß-g

al a

ctiv

ity

[RLU

*102

]/�

g p

rote

in

PMP22-lacZ*

*

c


49

Asterisks indicate statistical significance between values of transgenic and wildtypeanimals (Mann-Whitney U test, P<0.05). (b, c) Comparison of the endogenousPMP22 promoter with the transgenic PMP22 promoter activities in differenttissues. (b) About 3-fold more exon 1A than 1B-containing endogenous PMP22mRNA was detected in the sciatic nerve of 3-week old wildtype mice. In othertissues with PMP22 expression, mainly the 1B mRNA was found. (b) A similar ratioof 1A- to 1B-lacZ messages derived from the transgenic promoters could bedetected in the two -10/0kb PMP22 lacZ transgenic lines 44.2 and 48.4, although theabundance of LacZ mRNA was approximately two orders of magnitude lower thanthe endogenous PMP22 mRNA. Error bars represent the SD of all values (n=4-6)obtained from two independent experiments

5.7 -10/0kb PMP22 lacZ transgene regulation in Schwanncells after loss of axonal contact and in regeneration.

After a crush lesion of peripheral nerves, the axons distal to the injury degenerate.

During regeneration, they are replaced by regenerating axons growing out from the nerve

stump. During this process called Wallerian degeneration, the Schwann cells that lost

their axonal contact, initially dedifferentiate and proliferate, and finally remyelinate the

regenerated axons. Using this experimental paradigm, we examined whether the -10/0kb

PMP22 lacZ transgene also contains the necessary DNA elements for regulation of the

reporter gene by axonal contact as described for the endogenous PMP22 gene. Strong

downregulation of PMP22 had been seen within two weeks after a nerve crush followed

by an increase of PMP22 mRNA and immunoreactivity during remyelination (Snipes et

al., 1992; Suter et al., 1994). To this end, we performed nerve crush experiments in adult

-10/0kb PMP22 lacZ transgenic mice and analyzed ß-gal expression 9, 14 and 80 days

after the nerve crush in the distal part of lesioned nerves. These data were compared to

transgene expression levels of the corresponding nerves in the non-lesioned contralateral

nerve (Fig. 5-7a). Significant down-regulation of the transgene was detected nine days

after injury, even more pronounced after 14 days, followed by an upregulation to

expression levels similar to the contralateral nerve after remyelination (80 days post

crush).


50

Figure 5-7: Regulation of the -10/0kb PMP22 lacZ transgene during Walleriandegeneration and after regeneration. Temporal regulation of ß-gal expression aftera nerve crush (a) and after nerve transsection (b) of adult -10/0kb PMP22 lacZanimals. (a) Nine and 14 days after nerve crush, reduced ß-gal levels were observedduring Wallerian degeneration. After regeneration (80 days), ß-gal levels similar tothe contralateral nerve were detected. (b) 60 days after a nerve cut, the ß-galactivity of the distal part of the lesioned nerve was dramatically reduced comparedto the equivalent segment of the contralateral nerve, indicating that high ß-gallevels depend on axonal contact and myelination of the Schwann cells. Error barsrepresent the SD of the values for 3-4 animals. Asterisks indicate statisticalsignificance between values of lesioned and contralateral nerve (Mann-Whitney Utest, P<0.05).

By cutting peripheral nerves, reinervation distal to the cut is prevented. Using this

approach, we analyzed transgene expression in Schwann cells without axonal contact in

the distal part of the remaining nerve. Sixty days after the nerve cut, a strong

downregulation of the transgene was observed compared to the normal expression levels

in the non-lesioned contralateral nerve (Fig. 5-7b), indicating that axonal contact and

remyelination is necessary for full transgene activity.

To know whether there is a distal-proximal gradient of the transgene expression along a

nerve, we prepared in parallel additional homogenates of the proximal part of the

unlesioned nerve. Taken all measurement together, no significant difference could be

detected between the expression of the transgene in distal versus the proximal part of

unlesioned nerves (data not shown).

a

0

2 0

4 0

6 0

8 0

100

120

140

9 14 80

crushed nerve, distalcontralateral nerve

ß-g

al a

ctiv

ity

[% o

f co

ntr

alat

eral

ner

ve]

0

2

4

6

8

contra-lateral

distal(cut)

ß-g

al a

ctiv

ity

[RLU

*106

]/µ

g p

rote

in

days post crush

*

**

b


51

5.8 Sciatic nerves of PMP22 mutant animals show reducedß-gal levels

In a next step, we examined how the -10/0kb PMP22 lacZ transgene is regulated in

animal models for inherited peripheral neuropathies such as the Trembler mutant (Tr),

which has a point mutation in the coding region of the PMP22 gene (Suter et al., 1992a,

b), or in animals deficient for PMP22 (Adlkofer et al., 1995). These animals show

demyelination and incomplete remyelination without acute injury to the nerves (Sancho

et al., 1999, 2001). We analyzed the ß-gal activity in sciatic nerve homogenates of -10/

0kb PMP22 lacZ transgenic animals heterozygous for the Tr mutation compared to the ß-

gal activity in -10/0kb PMP22 lacZ transgenic animals without the Tr mutation. The

presence of the Tr allele led to a drastic reduction of the ß-gal activity in the sciatic nerve

at the age of three weeks (P21, Fig. 5-8a) and in adult animals (P60 and P90, Fig. 5-8a).

X-gal staining on teased nerve fibers of -10/0kb PMP22 lacZ transgenic Tr showed a

general reduced staining in all nerve fibers (data not shown).

In the same way, we examined the effects of PMP22 deficiency on -10/0kb PMP22 lacZ

transgene expression. -10/0kb PMP22 lacZ transgenic and PMP22-deficient mice were

analyzed at P21 and P60 and showed a significantly reduced ß-gal activity of the

transgene in the sciatic nerve compared to age-matched -10/0kb PMP22 lacZ transgenic

animals without PMP22 deletion (Fig. 5-8b). To check whether the neuronal expression

pattern of the transgene is changed in animals deficient for PMP22, whole mount

stainings of spinal cord and brainstem slices were analyzed, but no obvious changes were

observed (data not shown). X-gal staining on teased fiber preparations of the same

animals showed an inhomogeneous staining. Along the same fiber, some internodes

showed a strong X-gal staining whereas others stained only weakly (data not shown). It

has been previously reported that focal hypermyelination and myelin degeneration

occurs in the peripheral nervous system of adult PMP22 deficient mice (Adlkofer et al.,

1995). Therefore, the intensity of X-gal staining is likely to reflect whether the

corresponding Schwann cell is in a demyelinating or remyelinating phase. In addition,

the stripe-like accumulation of ß-gal at non-compacted myelin regions as seen in

wildtype animals (Fig. 5-3f), was not visible in either of the PMP22 mutants, indicating

an altered architecture of the myelin sheath as described by (Neuberg et al., 1999).


52

Figure 5-8: Activity of the -10/0kb PMP22 lacZ transgene in PMP22-mutant mice.ß-gal activity in sciatic nerve homogenates of the -10/0kb PMP22 lacZ transgene ona Tr mutant (a) or PMP22-deficient (b) background. A strong reduction of ß-galexpression in the sciatic nerve was observed in both PMP22 mutants at postnatalday 21 and in adult animals. ß-gal activities are shown in RLU*106. Asterisksindicate statistical significance between values of PMP22 mutant and wildtypeanimals (Mann-Whitney U test, P<0.05).

a

0

1

2

3

4

5

ß-g

al a

ctiv

ity

/ µg

pro

tein

wt Tr

P21 P60 P90

* * *wt Tr wt Tr

b

0

2

4

6

8

wt PMP22-/-

P21 P60

wt PMP22-/-

**

ß-g

al a

ctiv

ity

/ µg

pro

tein

PROMOTER DELETION ANALYSIS IN VIVO RESULTS PART III

53

6 PMP22 PROMOTER DELETION ANALYSIS IN VIVO

Part of this chapter will be the basis for a publication:

Marcel Maier, Francois Castagner, Philipp Berger and Ueli Suter (2003). Dissection of

the Peripheral Myelin Protein 22 (PMP22) Promoter in vivo Reveals a Late Myelinating

Schwann Cell Specific Element, manuscript in preparation

In a search for the cis-acting regulatory elements that control PMP22 expression in the

peripheral nervous system (PNS), we analyzed the expression of PMP22 - regulated

reporter constructs in transgenic mice. In a previous study, we started our analysis with a

ten kilobase fragment upstream of exon 2 of the PMP22 locus (Fig. 6-1a, cf. part II;

Maier et al., 2002a; referred to as -10/0kb PMP22 lacZ (PMP22-lacZ) construct, Fig. 6-

1b). In figure 1 and for the rest of the text, base +1 was defined at the A of the translation

start codon on exon 2. The 10 kb fragment contained promoter 1 preceding the

nontranslated exon 1A, promoter 2 preceding the alternatively used exon 1B and the first

part of exon 2 to the translation start codon. The alternative use of the two promoters

results in two mRNA that encode the same protein and differ only in their 5’ non-coding

region (Suter et al., 1994). We have shown previously, using quantitative RT-PCR, that

both promoters are active on this transgene since both transgenic transcripts could be

detected in addition to the two endogenous mRNA species in the siatic nerve (Maier et

al., 2002a). Our analysis showed that this 10 kb fragment contains regulatory elements

that reflect endogenous PMP22 expression in Schwann Cells during late myelination, in

PNS sensory and motor neurons, as well as during Wallerian degeneration and

remyelination after nerve crush.

To further dissect the cis-acting regulatory elements of PMP22 in the PNS, several

additional transgenic mouse lines with subfragments of this 10 kb element fused to the sh

ble-lacZ fusion reporter gene or to a heterologous hsp68 promoter fused to lacZ were

generated. The different reporter constructs tested as transgenes in vivo can be split

roughly into two groups. The first group of constructs was used to dissect the regulatory

elements responsible for the expression of the reporter gene in Schwann cells during


54

Figure 6-1: Reporter constructs used in transgenic mice to map PNS specificelements in the PMP22 promoter. (a) Diagram of the endogenous mPMP22 genomiclocus with promoter 1 preceding the nontranslated exon 1A and promoter 2 in frontof the alternatively used exon 1B. All numbers refer to the nucleotide +1, defined asthe "A" of the translation start codon on exon 2. (b-h) Constructs containingdifferent parts of the 10 kb region 5’ to exon 2 (-10/0kb region) were used to derivetransgenic mice. The -10/0kb PMP22-lacZ construct (Fig. 6-1b) was used in aprevious study (Maier et al., 2002a; PMP22-lacZ) and its activity reflects endo-genous PMP22 expression in Schwann cells during late myelination, in sensory andmotor neurons of the PNS, and during remyelination after a nerve crush. For thisstudy 5’ sequences extending from -10 kb to -4 kb (c), from -6.5 kb to -4 kb (d), orfrom -10 kb to -6.5 kb (e), were inserted at nucleotide -120 bp upstream of exon 2,which was in turn fused to a sh ble-lacZ fusion reporter gene. (f) To determine thefunction of the regulatory elements upstream of the core promoter 1/exon 1A afragment extending from -10 kb to -4.3 kb was fused to the heterologous corepromoter of the hsp68 gene. (g, h). By fusing -3/0 kb (g) or -4/0 kb promoter regions(h) to the sh ble-lacZ fusion reporter gene, the region between -4 kb and -3 kb wasanalysed. tss: translation start site, Sa: Sal I, K: Kpn I, St: Stu I, N: Not I

1A 1Bzeo lacZ pA

-10/0 kb PMP22 lacZ (PMP22lacZ)

1 kb

2K

-4.29

-4.18

-4.13

-1.66

N

-1.51

+1

1A 2-10/-4 kb Pro1 lacZ

zeo lacZ pA

-0.12

-3/0 kb Pro2 lacZ (PMP22-1B-lacZ)

hsp lacZ pA

-10/-4 kb hsp lacZ

2-6.5/-4 kb Pro1 lacZ 1A

zeo lacZ pA

1A 1B 2-4/0 kb Pro2 lacZ

zeo lacZ pA

1B 2

-2.96zeo lacZ pA

-10/-6.5 kb Pro1 lacZ 1A 2

zeo lacZ pA

-6.56

-4.01

-4.33

1A 1B 2 3 4

ATG

Promoter 1 Promoter 2 5

genomic mPMP22

a

b

c

d

e

f

g

h

-3.91

St

tss 1 tss 2

-10.04

+17.9 +25.5

Sa


55

myelination. They contained fragments of the 6 kb region upstream of exon 1A either in

front of the endogenous PMP22 promoter 1 (-10/-4kb Pro1 lacZ, -6.5/-4kb Pro1 lacZ and

-10/-6.5kb Pro1 lacZ, Figs. 6-1c, d, e) or in front of a heterologous promoter (-10/-4kb

hsp lacZ, Fig. 6-1f). The second group (-3/0kb Pro2 lacZ; PMP-1B-lacZ in Maier et al.,

2002a), -4/0kb Pro2 lacZ, Figs. 6-1g, h) contained fragments of the 4 kb region

immediately upstream of exon 2 including promoter 2 regions but lacking promoter 1.

6.1 Late myelination Schwann cell specific elements(LMSE) reside in the 6 kb DNA fragment upstream ofpromoter 1

Four out of the five mouse lines transgenic for the -10/-4kb Pro1 lacZ construct (Figs. 6-

3a) showed robust expression of the reporter gene in Schwann cells during postnatal

development in spinal nerves (e.g. sciatic nerve; Fig. 6-2) as well as in some cranial

nerves (e.g. trigeminal nerve, data not shown). Detailed analysis of the temporal and

quantitative expression of the reporter gene by whole mount X-gal staining and in

homogenates of the sciatic nerve at postnatal day P4, P12, and P30, revealed an

upregulation of the transgene beginning around P12 (Fig. 6-3b). This is rather late

compared to the timing previously reported for the endogenous PMP22 mRNA in rat or

mouse sciatic nerve (Maier et al., 2002a; Suter et al., 1994). A similar late expression

had been observed previously with the -10/0kb lacZ reporter gene (Maier et al., 2002a),

suggesting that late myelination Schwann cell specific elements (LMSE) are localized in

the -10/-4kb region.

X-gal staining of teased fiber preparations from sciatic nerves showed expression of ß-

gal in many Schwann cells associated with large caliber axons (Fig. 6-3c). The

percentage of ß-gal positive internodes varied between 40 and 80% in lines 828 and 830,

and between 5 and 15% in lines 827, 829, 831. Typical for the ß-gal staining of Schwann

cells on teased fiber preparations was the accumulation of cytoplasmic ß-gal in regions

with substantial Schwann cell cytoplasm, at the paranodes (Fig. 6-3c, filled arrowhead),

and in Schmidt-Lantermann incisures or perinuclear regions (Fig. 6-3c, open

arrowheads). Consistent with the finding that preferentially Schwann cells associated


56

Figure 6-2: Spatial expression patterns of transgenic reporter constructs in theperipheral nervous system. Presence (+) or absence (-) of reporter gene expressionwas examined using the histochemical ß-galactosidase assay, as described inMaterials and Methods. The analysis included samples from the period of maximalmyelin gene expression in postnatal development (P21 to P60). "+" was defined bystaining that contrasted with failure to stain in age-matched non-transgeniccontrols. Motor neuron and sensory neuron expression was analyzed on sectionsand on whole mount preparations of spinal cord tissue, sensory neuron expressionin addition by whole mount staining of dorsal root ganglia (DRG). Expression inSchwann cells was judged on whole mount X-gal staining and/or on teased fiberpreparations of the sciatic nerve. If no transgene transmission was achieved onlythe founder animal was analyzed (indicated with F). Footnotes: ° only observed in founder animal; (+): not consistently observed indifferent animals of the same line; a: plus 13 transgenic founder mice positive and 6

-10/0 kb PMP22 lacZ*

-10/-4 kb Pro1 lacZ

-3/0 kb Pro2 lacZ*

-10/-4 kb hsp lacZ

-6.5/-4 kb Pro1 lacZ

-4/0 kb Pro2 lacZ

motor sensory(DRG, sp.c.)

Schwann Cellslarge caliber

axonssmall caliber

axons

Neurons

Line (L) orFounder (F) #

L37.1 + + ++ +L44.2 ++ ++ +++ +L48.4 ++ ++ +++ +L45.2 (+) + + +a

L08.1 (h) d - - - -L95.2 (h) - - - -L10.2 (h) - - - - L12.2 (h) - - - -L19.1 (h) - - - -c

L818 - +° - -L819 d - + - -L820 d - +° - - F821 d - + - -L822 - + - -L823 d - + - -L824 (+) + + +L825 - +° - -L826 - - - -

F827 d - - ++ -L828 - - ++ +°L829 - + + -L830 - - ++ -L831 - - (+) -

F853 - - + +°

L858 d - (+) + -L860 - (+) + -L861 - - (+) -

L887 - - - - L907 - - - -L909 - (+) - -b

-10/-6.5 kb Pro1 lacZ ongoing


57

negative for ß-gal in peripheral nerves on tail sections; b: plus 5 transgenic foundermice negative on whole mount staining; c: plus 23 transgenic founder mice negativeon tail sections and 2 negative on whole mount staining; d: Line shows uniquepattern of expression outside the PNS, attributed to enhancer trapping at the site oftransgene insertion; h: ß-gal enzymatic activity measured in homogenates of sciaticnerve, heart, intestine, liver, lung, muscle or brainstem did not show any activityabove background; *: published in (Maier et al., 2002a)

with large caliber axons were positive for ß-gal, the ventral roots (which contain mainly

large caliber axons, arrows in Figs. 6-3d, e, f) were more intensively stained than the

dorsal roots (filled arrowheads in Figs. 6-3d, e, f) on whole mounts of the lumbar spinal

cord at P21 (Fig. 6-3d) and P90 (Fig. 6-3e). Furthermore, this expression of the transgene

is maintained in older animals (in mice of line 830 up to 13 months of age; data not

shown)

In contrast to mice with the -10/0kb lacZ transgene, four out of five mouse lines

transgenic for the -10/-4kb Pro1 lacZ construct showed no expression in sensory

neurons, as determined on whole mount X-gal stainings of dorsal root ganglia (DRG,

open arrowhead, Fig. 6-3f), and in motor neurons located in the ventral horn of the grey

matter in the spinal cord (asterisk in Fig. 6-3g), along the entire anterior-posterior axis.


58

Figure 6-3: -10/-4kb Pro1 lacZ transgenic mice show high ß-gal expression inSchwann cells of predominantly large caliber fibers in peripheral nerves and inspinal cord nerve roots. (a) schematic depiction of the -10/-4 kb Pro1 lacZtransgene. (b) ß-gal enzymatic activity in homogenates of the sciatic nerve atpostnatal day P4, P12 and P30 is indicated in the bar diagram. Whole mount X-galstaining of pieces of the corresponding sciatic nerve from -10/-4 kb Pro1 lacZtransgenic mice of line 830 are shown above the graph. Onset of ß-gal expression isaround P12 and increases dramatically during the late phase of myelination (P30).ß-gal activities are shown in relative light unit (RLU)*105 /µg protein. (c) X-galstaining on teased sciatic nerve of a three week-old mouse of line 828. Mainly largecaliber axons are enwrapped with ß-gal positive Schwann cells which show thetypical accumulation of the cytoplasmic ß-gal in regions with increased Schwanncell cytoplasm (open arrowheads: perinuclear, filled arrowheads: paranodal). (d, e)In whole mount X-gal staining of the spinal cord of 21-day (d) and 90-day old (e)transgenic mice of line 828, ß-gal expression is found in ventral roots (arrows) and

0

4

8

12

16

P12 P30

ß-g

al a

ctiv

ity

[RLU

*105

]/µ

g p

rote

in

days postnatal

d e

1A 2

-10/-4 kb Pro1 lacZzeo lacZ pA

a

P4

P90

P21

P90

f g

P21

b c

P21

**


59

to a lesser extent in the dorsal roots (arrowheads) of the lumbar spinal cord (rootmodality was identified by the site of insertion into the spinal cord). (f) Wholemount staining of dorsal root ganglion (DRG) with attached dorsal (filledarrowhead) and ventral roots (arrow). (g) Staining of crossection through thethoracic spinal cord of three week old transgenic mice of line 828. In four out of fivelines transgenic for the -10/-4kb Pro1 lacZ construct, no neuronal expression of ß-gal was detected either in the DRG where the cell soma of sensory neurons arelocated (f, open arrowhead) or in the ventral horn of the grey matter where the cellbodies of motor neurons can be found (g, asterisks). Scale bars: 50µm (c), 500µm(d,e), 100µm (f,g)

6.2 The LMSE confer Schwann cell specificity to the non-cell type-specific hsp68 promoter and are functionalindependent of core promoter 1 and exon 1A

To examine whether the 6 kb element contained in the -10/-4kb Pro1 lacZ construct had

functions characteristic of an enhancer and could act independently of core promoter 1,

exon 1A, and exon 2 sequences, we generated a construct with a slightly shorter segment

extending from -10 kb to -4.29 kb fused to the 0.3 kb minimal promoter of the heat shock

protein hsp68 (Fig. 6-4a). Thus, the core promoter 1 of PMP22 up to -120 bp from the

transcription start site as mapped by Suter et al. (1994), was replaced by the minimal

promoter of hsp68. This promoter contains a TATA-box, an SP1 recognition site, a

CCAAT box, and three heat shock response elements with no elements contributing to

tissue-specific expression (Kothary et al., 1988; Kothary et al., 1989). This promoter has

been succesfully used before to characterize distal Schwann cell enhancer elements

(Forghani et al., 2001; Mandemakers et al., 2000) or a retinoic acid response element

(Rossant et al., 1991) in vivo.

All four founder animals carrying this -10/-4kb hsp lacZ transgene expressed ß-gal

specifically in Schwann cells as determined by whole mount staining of sciatic nerves.

The percentage of ß-gal positive internodes varied between 44% for founder 853, 11%

and 12% for founder 858 and 861, respectively, and about 1% for founder 860. In the

founder animal 858 and in F2 animals of this line, robust expression of the transgene

could be detected in many Schwann cells mainly associated with large caliber fibers (Fig.


60

6-4c, d). Thus, this line was used to determine the temporal expression pattern during

early postnatal development in homogenates of the sciatic nerve at P4, P12 and P30 (Fig.

6-4b). A late upregulation of the reporter gene around P12 was detected. This timing is

comparable to that observed in animals transgenic for the -10/-4kb Pro1 lacZ construct

(Fig. 6-3b) and for the -10/0kb lacZ construct (Maier et al., 2002a). Robust ß-gal

expression was detected in Schwann cells by whole mount staining of the lumbar spinal

cord with attached roots (Fig. 6-4e). No expression was observed in the DRG where the

cell soma of sensory neurons are located (empty arrowhead in Fig. 6-4f), whereas the

nerves were strongly positive. In contrast to line 858, especially in the F1 and F2 animals

of lines 860 and 861, a highly mosaic expression of the transgene in Schwann cells was

observed. About 12% of the internodes were positive for ß–gal in the sciatic nerve of the

founder animal 861 (Fig. 6-4g), but this number decreased dramatically in later

generations (F1 generation: about 1%; Fig. 6-4h), with only some 5-10 positive

internodes per sciatic nerve remaining in the F2 generation (Fig. 6-4i).


61

Figure 6-4: The 6 kb LSME targets expression in the context of a heterologoushsp68 promoter. (a) The 6 kb LSME was ligated to a 0.3 kb hsp68 minimalpromoter and used to derive -10/-4kb hsp lacZ transgenic mice. (b) Bar diagramshowing the developmental expression of the -10/-4kb hsp lacZ transgene asdetermined by ß-gal enzymatic activity in homogenates of the sciatic nerve atpostnatal day P4, P12 and P30. Whole mount X-gal staining of pieces of thecorresponding sciatic nerve from transgenic mice of line 858 are shown above thegraph. Onset of ß-gal expression was around P12 and increased dramaticallyduring the late phase of myelination (P30). ß-gal activities are shown in relativelight unit (RLU)*105 /µg protein. (c, d) Teased fiber preparation of the sciatic nerveof the founder animal (c) and of a F2 animal of line 858 (d) with ß-gal positiveinternodal segments mainly detected on large caliber axons. (e, f) ß-gal can still bedetected in the attached dorsal (arrowhead) and ventral roots (arrow) by a wholemount X-gal staining of the lumbar spinal cord (e) or of the DRG (f) in the 11

Founder 858

f

e

hsp lacZ pA

F1 L861

F2 L861

-10/-4 kb hsp lacZ

a

b

c

F858

F2 L858

d

gP12 P30ß-g

al a

ctiv

ity

[RLU

*105

]/µ

g p

rote

in

days postnatalP4

F861

h

i

0

1

2

3

4

5


62

month old founder animal of line 858. No neuronal expression was detected inanimals with this transgene in the DRG (empty arrowhead in f). (g, h, i) In contrastto line 858, in F1 animals (h) of line 861 only few and in F2 animals (i) only singleinternodes positive for ß-gal could be detected on teased fiber preparations of thesciatic nerve.

6.3 The 4 kb sequence upstream of exon 2, includingpromoter 2, contains elements directing expression insensory neurons

As a first step to analyze promoter 2 associated regulatory elements we generated

transgenic mouse lines using the -3/0 kb sequence fused to the sh ble lacZ reporter gene

(-3/0kb Pro1 lacZ transgene (Maier et al., 2002a; PMP-1B-lacZ); Fig. 6-1g). This

construct showed cell-line specific expression in transfection experiments in the mouse

Schwann cell line MSC80 compared to the fibroblast NIH 3T3 cell line (cf. part I). Out

of 33 PCR-positive founders, only six mice showed weak and unspecific expression of ß-

gal in cross-sections of tail biopsies (data shown). These six founders and another

randomly chosen four founders were mated with B6D2F1 hybrid mice. None of the F1

animals showed expression in Schwann cells or neurons, nor were consistent significant

levels of ß-gal activity detected with the sensitive luminescence ß-gal assay in heart,

intestine, lung, muscle and brainstem (data not shown).

In the next step, we wanted to determine the expression pattern directed by the 4 kb

between exon 1A and exon 2 in vivo (-4/0kb Pro2 lacZ, Figs. 6-1h and 6-5a). In whole

mount X-gal staining of the DRG of three founder mice (lines 818, 820, 825) and four

out of five lines (lines 819, 821, 822, 823; Fig. 6-2), we observed large ß-gal positive

cells suggestive of sensory neurons (Figs. 6-5b, c; arrow). A detailed analysis of the

temporal expression pattern of mice from line 822 showed that the transgene was already

detected at postnatal day P4 in the DRG (Fig. 6-5b) and was still highly expressed in a

one year-old animal (Fig. 6-5e). The temporal expression of this -4/0kb Pro2 lacZ

transgene therefore recapitulates the sensory neuron expression pattern of the -10/0kb

PMP22 lacZ construct (Maier et al., 2002a).


63

Figure 6-5: Developmental expression of the -4/0kb Pro2 lacZ transgene showsspecific expression in sensory neurons. (a) Schematic view of the -4 kb Pro2 lacZtransgene. (b) Expression of lacZ detected at postnatal day P4 by whole mountstaining of the dorsal root ganglion (DRG, arrow) in transgenic mice of line 822. (c)At P21, lacZ expression is seen both in the DRGs (arrow) and in the dorsal columntract (gt) of the thoracic spinal cord. (d) On a cross-section through the thoracicspinal cord of a 90 day old mouse of line 821 ß-gal can be detected in addition in thedorsal horn of the grey matter (d, empty arrowheads). (e) Expression of ß-gal ismaintained in a one year-old animal (founder of line 822) and can be detected inhigh amounts in the processes of the sensory neurons by a diffuse X-gal staining inthe dorsal roots (dr) of the lumbar spinal cord. (f, g) On a teased fiber preparationof the sciatic nerve of the founder animal of line 822, the diffuse staining seen indorsal roots and in the sciatic nerve can be localized to a subset of myelinated fibers(f: phase contrast, g: bright field microscopy of the same fibers). The typicalaccumulation of ß-galactosidase reaction product in cytoplasmic compartments ofthe Schwann cells such as the paranode (arrowhead) is not seen and therefore theweak staining most likely derives from neuronal expression of ß-galactosidase. (h)Immunohistochemical staining of dissociated DRG from P4 animals. Neurofilament(NF, red) and ß-galactosidase (ß-gal, green) were detected in the same cells, gt:gracile tract, vr: ventral roots, dr: dorsal root, Scale bar: 200 µm (b-e), 20µm (f-h)

NF

ß-gal

1B 2-4kb Pro2 lacZ zeo lacZ pA

gt

L821P4

b c d

a

hgt

g

f

F822

drvr

vr

dr

e

L822

L822

P21

P30


64

Additional evidence that the ß-gal-positive DRG cells are neurons was provided by the

analysis of the ß-gal expression of the different cell types in vitro. DRG from P4 animals

were dissected, enzymatically dissociated and the cells placed in culture for three days.

Staining of the mixed cell population with antibodies against ß-gal and neurofilament

revealed a co-localization in cells with a neuronal morphology (Fig. 5-5h). In addition,

X-gal staining of sister plates showed ß-gal positive cells indentified by morphology as

neurons (data not shown).

Additional staining was observed in the higher expressing lines, such as 821 and 822, in

the dorsal column of the spinal cord as shown for a 90-day old F2 animal of line 822

(Fig. 6-5c). DRG neurons give rise to a peripheral extensions and to central axons. The

latter either terminate directly in the dorsolateral region of the spinal cord or ascend

ipsilaterally through the dorsal columns of the cord and terminate in the dorsal column

nuclei located in the lower medulla. Indeed, in line 821 and in the founder animal of line

822, additional staining was detected in the dorsal horn of the spinal cord (empty

arrowheads in Fig. 6-5d), as also observed previously in mice transgenic for the -10/0kb

lacZ construct (Maier et al., 2002a). In these high expressing lines 821 and 822 we

detected an additional weak and diffuse staining in dorsal roots (Fig. 6-5e), on whole

mount stainings of the sciatic nerve (data not shown), and in a subset of fibers in teased

fiber preparations (Fig. 6-f, g). However, the typical accumulation of the cytoplasmic ß-

gal staining in regions with increased Schwann cell cytoplasm such as at the paranodes

(arrowhead in Fig. 6-5f, g) was not observed, and the staining was always continuous

over several internodes. This pattern is that expected if expression occurs only in the

peripheral extensions of the sensory neurons and not in Schwann cells. This is also

consistent with the ß-gal activity we could detect in the extensions of the DRG neurons

of the dorsal columns of the spinal cord.

COMPUTATIONAL ANALYSIS OF PMP22 PROMOTERS RESULTS PART IV

65

7 COMPUTATIONAL ANALYSIS OF CONSERVED

PROMOTER ELEMENTS AND REPETITIVE GENOMIC

REGIONS

The availability of extensive sequence data for both the human and mouse genome has

provided the opportunity to identify regulatory elements using global sequence

alignments as functionally important sequences are often conserved during evolution

(Hardison et al., 1997; Mandemakers et al., 2000).

As a first step, we compared mouse and human sequences (Fig. 7-1a) using either the

VISTA program (Mayor et al., 2000)(see Material & Methods for details concerning

software and sequences) or the PIPMaker software (data not shown) to facilitate the

delimitation of regulatory elements in the whole PMP22 gene locus. Scanning with a 100

bp window, regions with more than 75% sequence identity were noted between -23 kb

and +40 kb (Fig. 7-1a). Homologous regions further upstream (> -30 kb) (data not

shown) most likely belong to the mouse homolog of TEKT3, which is the gene upstream

of PMP22 on the human sequence (Inoue et al., 2001). In general, regions with more

than 75% identity within the PMP22 gene locus were mainly found in the -10/0 kb

region but in addition also between exon 2 and +10 kb and upstream of exon 5 at the 3’

end of the gene. With the RepeatMasker program we annotated the mouse DNA

sequence for interspersed repeats and low complexity DNA sequences and visualized

them with the PIPMaker software as different boxes above the diagram (Fig. 7-1a,b,

legend in the figure; for review see Smit, 1996).

In a second step we focused our analysis on the -10/0 kb regulatory region that we have

characterized previously in vivo (Fig. 6-1a) (Maier et al., 2002a), and performed a

pairwise sequence alignments of three species (murine, human, rat) (Dubchak et al.,

2000) using the VISTA program. The human/rat alignment is very similar to the human/

mouse alignment (Fig. 7-1b) since the mouse and rat sequences are more than 75%

identical over almost the whole region (data not shown). As reported earlier, a high

nucleotide sequence conservation surrounding both transcription start sites and regions

rich in CpG dinucleotide around promoter 2 can be observed on the PIP (Fig. 7-1b)

(Suter et al., 1994; van de Wetering et al., 1999). We grouped the remaining regions of


66

Figure 7-1: Distribution of conserved DNA segments and repetitive DNA elementsin the mouse PMP22 gene locus combined with potential binding sites fortranscription factors. (a, b) Percent identity plot (PIP) generated using the VISTAalgorithm (Mayor et al., 2000) to compare the murine PMP22 genomic locus (AL592215) from -26 kb to +40 kb (a) or from -10.5 kb to +1bp (b) (horizontal axis; inrelation to the translation start codon (+1)) with the orthologous human sequences.The vertical axis indicates percent identity in a 100 bp window with a 31 bpresolution (a) or in a 50 bp window with a 15 bp resolution (b) of the plot. Regionswith >75% identical nucleotides are highlighted in gray. Note that the baseline is50%. The locations of coding exons (black rectangles), 5'- and 3'-untranslatedregions (U, open rectangles) and interspersed repeats (legend in the figure; Smit,1996) were identified using RepeatMasker software and are shown as differentboxes above the profile. Above the alignment for the -10/0kb PMP22 sequence (b)analyzed in the present study, conserved regions (CR) are numbered from 1 to 5. (c)


67

Potential binding sites for transcription factors for which a binding matrix haspreviously been defined, and which are known to play a role in myelination or to beexpressed in Schwann cells: Krox20/Egr-2, Egr-1, Tst-1/Oct-6, Pax-3, Peroxisomeproliferator-activated receptor (PPAR), Progesterone receptor binding site (PREBS), cAMP-responsive element binding protein 1 (CREBP), Brn-2/Pou3f2. (d)Alignment with the PMP22 promoter driven constructs analyzed in the presentstudy. Sequencing of the -10/0kb PMP22 DNA segment used for the cloning of ourconstructs revealed that the repetitive region from -8.0 to -7.46 (in a, b) is missingcompared to the published mouse PMP22 sequence (AL 592215) (indicated with adotted line).

high identitiy of the human/mouse alignment into five conserved regions (CR, Fig. 7-1b).

In CR 1 two peaks of high identity were found, whereas CR 2 and CR 3 mark regions

with an overall identity above 50% but without larger regions above 75%.

The -10/0kb sequence was screened for potential binding sites with the Mat Inspector

program (Quandt et al., 1995). Among the 1080 putative elements identified, we selected

the specific transcription factors for which a binding matrix once had previously been

defined (for references see below and Material and Methods), and which were described

to be involved in myelination or to be expressed in Schwann cells (Wegner, 2000a; b).

The sites chosen are: Krox20/Egr-2 and Egr-1 (Swirnoff and Milbrandt, 1995), Tst-1/

Oct-6 (He et al., 1991), Pax-3 (Chalepakis and Gruss, 1995), Peroxisome proliferator-

activated receptor (PPAR) (Palmer et al., 1995), Progesterone receptor binding site (PRE

BS) (Nelson et al., 1999), cAMP-responsive element binding protein 1 (CREBP) (e.g.

Benbrook and Jones, 1994; Paca-Uccaralertkun et al., 1994), Brn-2/Pou3f2 (He et al.,

1989; Li et al., 1993) (Fig. 7-1c). Interestingly, many potential binding sites for Oct-6,

Brn-2, Egr-1 and Egr-2 are found in the -10.5/-9kb region although no longer stretches of

sequence similarity were found between the human and mouse sequence. Nevertheless,

we found potential binding sites for these transcription factors also on the corresponding

human sequence. In conserved regions, a potential binding site for Egr-2 can be found in

CR2 and for the progesterone receptor in CR1 and CR4.

In Fig. 7-1d the results from this computer screening are aligned to the PMP22-promoter

constructs analysed as transgenes.

PMP22 GENE REGULATION DISCUSSION & OUTLOOK

69

8 DISCUSSION AND OUTLOOK

8.1 PART I: MYELIN GENE REGULATION IN VITRO

Commitment, differentiation and maturation of neural cells are dependent on complex

programs that determine specific patterns of gene expression. Thus, the elucidation of the

regulation of neural gene expression will provide important information on the cellular

mechanisms involved in the differentiation and maturation of the nervous system. In this

context, I have analyzed the regulation of the PMP22 gene. The gene is of particular

interest, because of its gene-dosage sensitivity, leading to hereditary peripheral

neuropathies, and because of its pivotal role in Schwann cell biology and myelination.

I tested the suitability of different cell culture systems for reliable studies of PMP22 gene

expression. I started with a classical in vitro promoter deletion analysis, by transfection

of constructs containing different fragment of the 10kb region upstream of exon 2 fused

to a lacZ reporter gene. I found that promoter 1 of PMP22 was active only at very basal

levels in the mouse Schwann cell line MSC80 or in cultured Schwann cells. On the other

hand, promoter 2 derived PMP22 expression was found in considerable amounts in those

cells. This is probably due to the fact that cultured Schwann cells without coculturing

with neurons do not myelinate in vitro, so that the Schwann cells do not integrate axonal

signals into their transcriptional regulation. This is in contrast to the in vivo situation in

the peripheral nerve, where promoter 1 derived 1A-PMP22 message is predominant

during myelination (Suter et al., 1994). Indeed, PMP22 expression in cultured Schwann

cells can be increased by addition of the adenylate cyclase activator forskolin, which can

mimic under certain conditions some aspects of axonal contact of Schwann cells (e.g.

Lemke and Chao, 1988; Trapp et al., 1988). This confirms the observation that promoter

1 is the myelin-associated promoter and very likely responsible for the expression levels

of PMP22 during myelination, which are much higher than the promoter 2-derived

constitutive expression levels in non-myelinating Schwann cells or other cell types.

Consequently, in my opinion, it is not reasonable to study promoter 1-associated


70

expression of PMP22 in cultured Schwann cells by studying, for example, promoter

deletion constructs of promoter 1 in transfections. Nevertheless, an extensive promoter

deletion study was performed by Hai et al., (2001) in the RT4-D6P2T rat schwannoma

cell line (Imada and Sueoka, 1978; Yamada et al., 1995). In this cell line, they detect

some expression of the promoter 1 associated 1A-PMP22 and high expression levels of

1B-PMP22 transcripts in RNase protection assays after addition of high forskolin

concentrations (50µM). In a promoter deletion study with constructs ranging from -3451

bp to -43 bp of the human sequence (relative to the transcription start site 1 on exon 1A,

Suter et al., 1994) directly fused to the luciferase reporter gene, they detected expression

levels in the range of the SV40 promoter activity (if the SV40 construct was also

transfected in equimolar amounts). The relative expression levels increased slightly with

each additional 5’ deletion up to -105 bp. The last deletion to -43 bp led to a complete

loss of reporter gene activity. My present interpretation of these results is, that they

reflect basal promoter activity that increases with each additional deletion of – at least in

these cells – negative acting ‘myelin specific elements’. This basal promoter activity is

abrogated with the deletion of the obviously essential -105/-43bp promoter region, which

is just upstream of the core promoter 1. Note that in vivo this sequence can be replaced

by a hsp68 minimal promoter without loss of cell type specificity (cf. part III). On the

other hand the differences in the results might also be due to the use of the human

promoter sequence, to the different cell line used, or due to the direct fusion of the

reporter gene to exon 1A so that a splicing event is not required for correct expression.

Another result from my promoter deletion analysis was that expression levels of the

PMP22 promoted reporter constructs were considerably higher in MSC80 cells

compared to NIH3T3 cells. For this comparison the SV40 promoter expression levels

serve as a standard with the assumption that the SV40 promoter is expressed at similar

levels in both cell types. If these different relative expression levels indeed can be

confirmed by measurement of the absolute expression levels, one could conclude that the

MSC80 cells contain certain factors which increase the low levels of PMP22 expression

and which are not abundant in NIH3T3 cells. It would be very interesting to identify

these factors which are differentially expressed in only one cell line. This could be done

for example on RNA levels using microarray technology or on protein levels using 2D-

gel electrophoresis and other proteomic approaches.


71

Another cell culture system in which PMP22 expression has been studied involves in

vitro myelination of DRG neurons by Schwann cells (Pareek et al., 1997). Unfortunately,

in vitro myelination systems of the PNS, especially with mouse tissue, are not very

reliable and nearly as time-consuming as in vivo experiments. In addition from my own

experience, they are not established to the degree that additional manipulations or

modifications are tolerated or can be reliably evaluated in a quantitative manner.

Therefore I did not present my initial in vitro myelination experiments to characterize the

-10/0kb PMP22 lacZ transgenic mice in the present thesis.

Since we did not have a cell culture system in which both promoters of PMP22 were

reliably expressed, I shought another setting in which PMP22 expression could be

induced e.g. by exogenously altered expression of transcription factors. This was done

successfully for the myelin protein zero (MPZ) in cotransfection experiments (Monuki et

al., 1990; Zorick et al., 1999) or with doxycycline induction of transcription factors in

N2A cells (Peirano et al., 2000). Unfortunately, I did not observe any upregulation of

PMP22 promoter-directed reporter gene expression by cotransfection of various

transcription factors (see chapter 4.1) nor did I see a co-regulation of PMP22 with P0 by

Sox10 in N2A cells (see chapter 4.3). At least in this setting, PMP22 seems not to be

regulated by Sox10. This is astonishing, since these two myelin genes have a comparable

temporal and spatial expression pattern, at least postnatally in myelinating Schwann

cells.

One may speculate why these systems are not working, but it must be realized that

complex interactions and combinations of known and of probably still unknown

regulatory factors finally lead to the correct temporal and spatial expression in a complex

organ like the PNS. In addition the cotransfected transcription factors are usually

strongly overexpressed and may therefore “overload” the physiology of the cell (e.g. the

RNA splicing machinery), and additionally required endogenous factors may not be

available in sufficient amounts.

Even so, Nagarajan and coworkers found a way to circumvent some of these problems.

Infection of cultured Schwann cells with adenovirus expressing Krox20 led to an

upregulation of myelin genes, with steady-state mRNA levels rising between 5-fold


72

(MBP) and 60-fold (for MPZ)(Nagarajan et al., 2001). This means that transcription of

myelin genes in their endogenous chromosomal context can be induced by the

overexpression of Krox20. I could reproduce this up-regulation of myelin genes, at least

in the absence of serum. The presence of 10% FCS in the cell culture medium prevented

a strong upregulation of PMP22. Possibly mitogens in the serum prevent the Schwann

cells from attaining a pro-myelinating-like phenotype and concomitant upregulation of

PMP22/growth arrest specific gene 3 (gas-3) (Zanazzi et al., 2001).

The expression levels of the myelin genes in cultured Schwann cells overexpressing

Krox20 are still much below the expression levels in myelinating Schwann cells. This

points to the fact that additional factors or signals are required for increased expression

of the myelin genes. A screening for those additional components could be performed,

for example by expression cloning, or also by treatment of Schwann cells with

membrane- and soluble fractions of DRG neurons since neuronal contact is described to

upregulate PMP22 expression. In any event, Krox20-infected Schwann cells may allow a

dissection of the regulatory network involved in the initiation of myelin gene expression

in vitro by studying co-regulated or target genes of Krox20. I tried to confirm by

quantitative RT-PCR the up-regulation of certain candidates previously identified in gene

expression profiling experiments (Araki et al., 2001; Nagarajan et al., 2001). I could

confirm only the upregulation of the POU domain transcription factor Brn-2 by Krox20.

Unfortunately, I could not confirm the upregulation of our additional candidates. This

could be due to slightly different cell culture conditions, a PCR which is not sensitive

enough, or to false positives in the gene chip analysis (the data presented in Nagarajan et

al., 2001 seem to derive from a single chip hybridisation experiment).

Indeed, in the meantime it has been shown that Brn-2 has an expression pattern similar to

that of Oct-6 during Schwann cell myelination and after nerve crush (Sim et al., 2002).

This makes Brn-2 an interesting candidate for further studies. That could be used to

dissect this regulatory network of Krox20-induced myelin gene regulation. For example

adenoviral overexpression in cultured Schwann cells could be used to ask whether Brn-2

is able to directly regulate myelin gene expression or whether it has a cooperative effect

if coexpressed with Krox20. Finally the target genes of Brn-2 could be identified using

microarray technology, as has been done for Krox20 (Nagarajan et al., 2001). New

candidates for further analysis might thus be discovered. Another possibility would be to


73

identify the pathways that are involved in the initiation of the PMP22 gene expression by

the addition of specific inhibitors, as has been done by Awatramani et al., (2002). Once

interesting pathways are identified, specific candidates could be blocked, for example by

RNA interference experiments.

8.2 PART II: REGULATORY ELEMENTS ON THE -10/0KB

PMP22 LACZ TRANSGENE

The development and proper function of peripheral nerves in vertebrates depend on

intimate interactions and continued signalling between Schwann cells and the associated

axon(s). In recent years much progress has been made in identifying components of cell-

cell interactions necessary in early stages of peripheral nerve development (Mirsky and

Jessen, 1999; Mirsky et al., 2002). In contrast, little is known about extracellular signals

and intracellular signalling pathways that initiate and regulate myelination. Earlier work

has indicated that the myelination program of Schwann cells is under the control of the

associated axon and correlates with axonal diameter (Aguayo et al., 1976a; b; Voyvodic,

1989). Whatever the exact nature of these signals might be, finally they must be relayed

to the Schwann cell nucleus where transcription factors coordinate the regulation of sets

of genes.

To start the dissection of the coordinate regulation of PMP22 in vivo I have produced

transgenic mice carrying both the lacZ reporter and the zeomycin resistance (sh ble)

genes driven by PMP22 promoter 1 and promoter 2 plus additional potential regulatory

domains in the 10 kb region 5` to the coding sequence of the PMP22 gene. I examined

the role of this region in the spatiotemporal transcriptional regulation of PMP22 with a

lacZ reporter gene in vivo during development, regeneration, and in animal models of

hereditary peripheral neuropathies. Consistent with the endogenous expression of

PMP22, by far the highest expression levels of the -10/0kb PMP22 lacZ transgene were

observed in Schwann cells of peripheral nerves during myelination (Haney et al., 1996;

Snipes et al., 1992; Spreyer et al., 1991; Welcher et al., 1991). Dramatic upregulation of


74

the transgene was observed around P8-10 during the most active phase of myelination.

Upregulation of the endogenous PMP22 gene, however, starts already with the initiation

of PNS myelination shortly after birth. The delayed upregulation of the transgene may

have various causes. First, reliable quantitative detection of the very low lacZ mRNA

levels in the sciatic nerve in the first postnatal days was not possible probably due to a

general low abundance of the lacZ mRNA as it has also been observed in other studies

(Feltri et al., 1999; Wight et al., 1993). This might be the result of decreased mRNA

stability or different post-transcriptional regulation of the lacZ mRNA compared to the

PMP22 mRNA (Bosse et al., 1999). Alternatively, regulatory elements for the early

PMP22 transcription could be missing within the 10 kb 5`-flanking region used in our

reporter study. This interpretation offers the hypothesis that the upregulation of PMP22

during myelination (and possibly also of other genes encoding myelin components) is

controlled on the molecular level by various distinct factors. On a speculative view, one

might envisage that some specific factors are important at myelination initiation while

others contribute to the exceptionally high gene expression that is required during the

peak of myelination or later during adulthood in myelin maintenance (cf. part III).

Annother possibility would be that the PMP22 transgene might interfere with

endogenous PMP22 expression, but I have not observed any evidence for this. In

particular, I have also not observed signs of delayed myelination as analysed at postnatal

day P4 on semithin sections through the sciatic nerve (data not shown). Furthermore the

ratio of 1A-LacZ to 1B-lacZ mRNA (transgenic) was similar to the ratio of 1A-PMP22

to 1B-PMP22 mRNA (endogenous) indicating that no important regulatory elements are

missing for similar relative promoter activities of the two transgenic promoters compared

to the two endogenous promoters at postnatal day 21.

The expression of the -10/0kb PMP22 lacZ transgene after nerve lesions and during

remyelination follows the spatio-temporal expression pattern of the endogenous PMP22

gene. These experiments strongly suggest that PMP22 transcriptional regulation in

Schwann cells is dependent on axonal contact and that full activity of the PMP22

promoters is dependent on myelination (Gupta et al., 1993; Spreyer et al., 1991). The

observations from cell culture experiments with -10/0kb PMP22 lacZ transgenic mice


75

are consistent with this notion: Schwann cells isolated from postnatal sciatic nerves

showed only very low ß-gal activity (data not shown).

To analyse the regulation of the -10/0kb PMP22 lacZ transgene in settings with impaired

myelination, demyelination and only limited remyelination (without actively severing

the axon, but see also Sancho et al., 1999), mice carrying the transgene were crossed

with Trembler (Tr) and PMP22-deficient mice. The drastically reduced levels of the -10/

0kb PMP22 lacZ transgene expression in the sciatic nerve of the resulting animals

(Fig.5-8) was likely due to the presence of fewer myelinating Schwann cells (Adlkofer et

al., 1995; Suter et al., 1992a), consistent with the reduced levels of PMP22 mRNA levels

in Tr mice (Garbay et al., 1995). Thus, the mutant Schwann cells are not capable of

maintaining the normal program regulating myelin gene expression, possibly due to

impaired axon-Schwann cell signalling (Sancho et al., 1999) or to an intrinsic failure to

differentiate.

The detection of the reporter gene product in sensory and motoneurons of peripheral and

cranial nerves confirms the expression of PMP22 mRNA found in motor neurons

(Parmantier et al., 1995) and in the DRG during early postnatal development (Parmantier

et al., 1997). These findings indicate that the control elements required for neuronal

expression are contained within the -10/0kb PMP22 lacZ transgene and are consistent

with the detection of low level PMP22 immunoreactivity in the DRG and the dorsal horn

of the spinal cord (De Leon et al., 1994).

The expression of the transgene during embryonic development was roughly consistent

with the previously described expression pattern of PMP22 mRNA (Baechner et al.,

1995; Parmantier et al., 1997). However, the relevance of the prominent ß-gal staining in

the outer ear, the limb muscle, and in the ventricular epithelium of the rhombencephalon

remain to be determined.

One of the potential benefits of this study was the generation of an experimental tool that

could be used to direct foreign gene expression reliably in Schwann cells and their

associated neurons to examine myelination and dysmyelination. The PMP22 gene


76

regulatory region used here proved to be very consistent in its expression pattern between

different lines, and was robust in the number of expressing founder animals. Importantly,

I could not detect consistent transgene expression in postnatal non-neural tissues such as

small intestine or lung, despite the fact that low levels of PMP22 mRNA had been found

before (Fig. 6-6; Patel et al., 1992; Taylor et al., 1995). This finding indicates that, on one

hand, the regulatory elements required for this non-neural PMP22 expression are

probably missing on our transgene (or that I am below the detection limit of the assay).

On the other hand, it makes this PMP22 regulatory region an even more valuable tool

due to its strict tissue specificity.

Furthermore, taking into account the strong upregulation of the reporter gene during

myelination, the -10/0kb PMP22 lacZ animals are a valuable tool to investigate potential

myelin gene regulatory factors that might also be involved in the pathogenesis of

hereditary peripheral neuropathies. Such an approach is currently being followed in

collaboration with R. Melcangi and his group. Currently, they are analysing the effect of

progesterone derivatives on the transgene expression and on PMP22 mRNA levels in

vivo. Previous in vitro and in vivo experiments have shown that certain progesterone

derivatives are able to influence PMP22 and P0 gene expression (Chan et al., 2000;

Desarnaud et al., 1998; reviewed by Magnaghi et al., 2001; Melcangi et al., 1999; 2001).

Indeed, initial experiments show, that the -10/0kb PMP22 lacZ transgene can be

upregulated by repeated administration of tetrahydroprogesterone (R. Melcangi, personal

communication). Since some conserved potential progesterone receptor binding sites are

found in the -10/0kb sequence (cf. part III), progesterone derivatives potentially could act

directly on the PMP22 gene promoter. To narrow down the sequence of the progesterone

responsive regions of the PMP22 promoter, different reporter transgenes (cf. part III)

could be tested for their response to progesterone derivatives. Since the reporter gene

allows a screening for different PMP22 regulatory factors in vivo, further studies will

focus on potential pathways involved in the progesterone-mediated PMP22 and PMP22

reporter transgene up-regulation in vivo.


77

8.3 PART III & IV: PMP22 PROMOTER DELETIONANALYSIS IN VIVO

The results of the detailed characterization of the -10/0kb PMP22 lacZ transgene show a

specific regulation and expression in Schwann cells and in a subset of peripheral

neurons. This justified further effort in delineating PNS specific elements.

To this end I generated a first group of transgenic mice containing different fragments of

the 6 kb region 5’ of Promoter 1 fused to both a lacZ reporter and a zeomycin resistance

(sh ble) gene. The transgenes contained the regulatory elements sufficient for expression

of the reporter gene during myelination in Schwann cells. In contrast to the expression

pattern of the previously characterized -10/0kb PMP22 lacZ transgenic mice I did not

observe neuronal expression of the reporter gene (cf. part II, Maier et al., 2002a). With

the fusion of this 6 kb fragment to a lacZ reporter gene with a minimal promoter of the

hsp68 gene, I showed that these cis-acting elements have enhancer-like properties, since

they are sufficient for expression of the reporter gene independent of the core promoter 1

and sequences of exon 1A. I analysed a second group of transgenic mice which

contained a 3 kb or a 4 kb fragment 5’ to exon 2 fused to the sh ble lacZ reporter gene.

Although I did not observe expression in the PNS with the 3 kb fragment, the 4 kb

fragment showed expression in sensory neurons of the PNS.

I observed strong upregulation of the transgenes with the myelin-specific Schwann cell

elements (-10/-4kb Pro1 lacZ and -10/-4kb hsp lacZ) around postnatal day 12, during the

most active phase of myelination and of expression of the endogenous PMP22 gene

(Haney et al., 1996; Snipes et al., 1992; Spreyer et al., 1991; Welcher et al., 1991). In

comparison with the endogenous PMP22 gene which begins upregulation with the

initiation of PNS myelination shortly after birth, the reporter constructs have a delayed

upregulation. This phenomenon was already observed with the 4 kb longer -10/0kb

PMP22 lacZ construct (Maier et al., 2002a) and could be due to missing regulatory

elements responsible for the early induction of PMP22. This interpretation offers the

exiting hypothesis that different regulatory mechanism are involved in different phases of

myelination and that these late myelinating specific pathways converge on this regulatory

DNA element. Consequently I termed this DNA element a late myelinating Schwann cell


78

specific element (LMSE), which I located in this study within the 6 kb region in front of

promoter 1. In the previous part (part II) I showed that the -10/0kb region in the -10/0kb

PMP22 lacZ construct contains the regulatory regions needed to reflect the

spatiotemporal expression of the endogenous PMP22 also after nerve lesions and during

remyelination. I have shown that the Schwann cells upregulate the transgene specifically

during myelination and have only very basal expression of the transgene without axonal

contact. So far I did not observe any difference in spatiotemporal expression of the -10/-

4kb Pro1 lacZ construct compared to the -10/0kb PMP22 lacZ transgene. In accordance

with this observation, the -10/-4kb Pro1 lacZ transgene is still highly expressed in one

year old animals. Therefore I assume that the LMSE contain the regulatory regions and

may bind transcription factors specifically involved in myelin maintenance and probably

also regeneration.

So far little is known about transcription factors specifically involved in remyelination or

myelin maintenance. Presumably these factors would be expressed at high levels in

myelinating Schwann cell. Recently, the POU domain transcription factor Brn-5 has

been shown to be expressed in increasing amounts in late Schwann cell development

(Wu et al., 2001), with an expression pattern inverse to that of SCIP/Oct-6. Therefore

Brn-5 could be involved in these regulatory mechanisms. So far no regulatory elements

have been described to be responsible specifically for expression in the late phase of

myelination. Only the previously characterized MBP Schwann cell enhancer (SCE1) has

been described to maintain expression of the reporter gene also in old adult animals, at

least in some lines (Forghani et al., 2001). For the myelin specific element (MSE) of the

Krox20 gene, which leads to specific expression in myelinating Schwann cells, it would

be very interesting to follow this aspect of expression since it has not been addressed so

far in old adult animals (Ghislain et al., 2002). However, the existence of elements

responsible for early- versus late expression of myelin genes has been discussed in the

case of oligodendrocyte-specific expression of MBP promoter-lacZ reporter genes. The

expression of certain constructs declined in old adult animals, whereas it was maintained

in constructs with different regulatory sequences (reviewed by Ikenaka and Kagawa,

1995). In the PNS, a similar phenomenon has been observed for CNP promoted lacZ

reporter gene expression, which only partially recreated endogenous CNP expression


79

since lacZ was only detected at early stages of development, but not in older adult mice

(Gravel et al., 1998; reviewed by Wegner, 2000a). Since not many transcription factors

involved in this type of regulation are known, a combination of the obtained

spatiotemporal expression pattern of the LMSE reporter constructs with whole genome

expression studies in the peripheral nerve, selecting for co-regulated genes with similar

expression profiles, may help to reveal new candidates involved in the late phase of

myelination, in myelin maintenance or regeneration.

The fact that the LMSE is sufficient for specific expression in Schwann cells in front of a

minimal hsp68 promoter in the -10/-4kb hsp lacZ construct shows that the LMSE has

enhancer-like characteristics and that the endogenous promoter 1 and exon 1A sequences

are not necessary for tissue specific expression since they can be replaced by the

unspecific hsp68 minimal promoter. By comparing at least three different lines of a given

construct I have tried to average out effects due to different expression levels of the

randomly integrated transgene, which depend on copy number and in many cases also on

the integration site. From such comparison, I observed one difference between the spatial

expression of ß-gal on teased fiber preparation of mice transgenic for the -10/-4kb Pro1

lacZ construct with animals transgenic for the -10/-4kb hsp lacZ. It was more difficult to

establish stable transgene-expressing lines with the latter construct since two out of three

lines showed a highly mosaic expression (variegation) in F1 and F2 generations. In

addition, Igot the impression that mosaic expression also of the -10/-4kb Pro1 lacZ was

more common than with the -10/0kb PMP22 lacZ transgene. The interpretation of these

observations may involve different explanations and mechanisms which are not clearly

distinguishable. Traditionally, enhancers have been thought to function by increasing the

rate of transcription initiation from a linked promoter. In recent years, it has been shown

that in some cases enhancers not so much influence the rate of transcription initiation, but

instead increase the chance that a linked promoter is activated at all. In this probabilistic

model, enhancers are thought to function through a mechanism that involves

modifications to the local chromatin configuration or relocation to an active centre within

the nucleus (Fiering et al., 2000; Hume, 2000). This model predicts that an enhancer

increases the percentage of cells in a population expressing the gene. In transgenic mice

experiments, such a mechanism might explain the often observed variegated expression


80

of the transgene (Elliott et al., 1995; Milot et al., 1996). Consequently, the loss of

enhancer activity could lead either to a smaller number of Schwann cells expressing the

reporter gene rather than a generally lower expression of the reporter gene in internodes

resulting in a less intense staining. So far intensities of different ß-gal stainings can be

compared, this is in fact what I observed, both, in the -10/-4kb Pro1 lacZ compared to the

-10/0kb PMP22 lacZ and in the -10/-4kb hsp lacZ compared to the -10/-4kb Pro1 lacZ

transgenes. In the first case, one can speculate that the -10/0 kb region acts as a stronger

enhancer than the -10/-4kb LMSE in Schwann cells by modulating the chance that the

linked promoter 1 is activated. Alternatively, one can argue that the LSME acts on both

PMP22 promoters instead of just promoter 1. In the second case, one may speculate that

some enhancer activity was lost due to the replacement of the endogenous promoter 1,

exon 1A and exon 2 sequences by the hsp68 promoter (cf. Fig. 7-1). Alternatively, one

can not exclude an effect due to the different reporter cassettes used. However, finally we

do not have enough data and the appropriate system to determine which of the

mechanisms discussed above dominate. The effects probably overlap, especially due to

the removal of the enhancer from its native context (Graubert et al., 1998; Sutherland et

al., 1997). These mechanisms can be taken apart to a certain degree with a system as

introduced in the section below.

A reproducible integration of the transgene always as a single copy at the same defined

genomic locus would have the advantage that absolute expression levels of different

transgenes can be compared (Guillot et al., 2000; Misra et al., 2001). This can be

achieved in ES cells, modified with a partially deleted HPRT locus. A transgene can be

inserted into the genome with a construct that allows reconstitution of the HPRT locus at

the same homologous recombination site in multiple transformants. This system may

allow to distinguish the different aspects of enhancers as mentioned above. In

collaboration with the groups of Alan C. Peterson and G. Jackson Snipes ES cells

containing a -21/0kb lacZ and a -11/0kb lacZ constructs are currently being tested for the

spatiotemporal expression pattern of the reporter gene in mice. Interestingly, preliminary

analysis of the -21/0kb lacZ construct indicates that there is no high expression levels of

ß-gal in sciatic nerves (G. Snipes, personal communication). The analysis of the -11/0kb

lacZ construct in mice will show whether this is due to the integration of a single copy of


81

the reporter gene, which might be expressed at levels too low to be detected, or whether

this is due to additional inhibitory elements upstream of the -10/0 kb fragment analysed

in the present study. The most rigorous way to test the function of these transcriptional

control elements in their native context would require specific deletion by homologous

recombination in embryonic stem cells. Such deletion could result in the generation of a

hypomorphic allele in case of the deletion of a cell-type specific enhancer (for example

see Ghazvini et al., 2002 with the Oct-6∆SCE/∆SCE mice). For PMP22, it would be of

interest to see whether mice with a deletion of the LMSE develop a myelination

phenotype caused by low expression of PMP22. This is likely since a reduction of

PMP22 expression leads to HNPP in humans (Chance et al., 1993) and to a HNPP-

comparable phenotype in mice heterozygous for the PMP22 deletion (Adlkofer et al.,

1997a).

To continue the dissection of the regulatory network of PMP22, it would be interesting to

assess whether the Oct-6∆SCE/∆SCE mice (Ghazvini et al., 2002) show delayed induction

of PMP22 mRNA during their delayed myelination. This is likely since Ghazvini et al.,

(2002) have shown that also Krox-20 and P0 expression is delayed in these mice. In this

case, it would be of interest to see if these animals completely compensate for the lower

PMP22 levels in old adult animals. This would provide some evidence that a different set

of regulatory factors is involved in myelin maintenance, or at least that Oct-6 is not the

rate-limiting transcription factor. Similar questions could be addressed with a tamoxifen-

inducible creERT2 transgene expressed in adult myelinating Schwann cells (e.g. a PLP

creERT2, a P0Cx32 creERT2 or a -10/0kb PMP22 creERT2 which all are under

construction) combined with a “floxed” allele for a transcription factor such as the

Krox20 (Taillebourg et al., 2002). These mice would recombine the Krox20 allele in

Schwann cells upon tamoxifen induction, which then should lead to a tissue-specific

ablation of Krox20 in adult myelinating Schwann cells.

Most mouse lines with the -10/-4kb hsp lacZ or -10/-4kb Pro1 lacZ transgene show

expression mainly in Schwann cells associated with large calibre fibres. This is in

contrast to the -10/0kb PMP22 lacZ construct, where ß-gal positive Schwann cells are

also associated with middle and small calibre fibres, albeit with clearly reduced ß-gal


82

levels. In agreement with this observation, the ventral roots of the lumbar spinal cord,

which contain only large calibre motor axons were more strongly stained than the dorsal

roots which contain the sensory axons (Figs. 6-3d, e). This could be explained by lower

expression levels of the transgenes, which would not be detectable in Schwann cells of

small calibre fibres. The latter already showed lower expression levels in -10/0kb PMP22

lacZ mice (Maier et al., 2002a). Alternatively this finding might be due to the lack of

additional positive regulatory elements outside the 6 kb LMSE element. If correct, this

may suggests a novel level of heterogeneity among Schwann cells that would be

correlated with the fiber diameter of the associated axon. Such molecular differences

between Schwann cells associated with motoneurons versus sensory neurons have been

described before (Martini et al., 1994).

The majority of transgenic mice lines carrying the -4/0kb Pro2 lacZ construct show

expression of the transgene specifically in sensory neurons Since the -3/0kb Pro2 lacZ

construct does not give rise to expression in the PNS (Maier et al., 2002a), it can be

concluded that regulatory elements located between -4 and -3 kb are necessary for

expression of the transgene in sensory neurons. In contrast to the neuronal expression in

the PMP -10/0 kb lacZ transgenic mice, no expression in motor neurons was observed,

suggesting that expression in motor neurons probably needs additional regulatory

elements located further upstream.

At this point it can be asked whether promoter 2-associated expression of PMP22 alone

is responsible for neuronal expression of PMP22, since in the CNS mainly 1B-PMP22/

SR13 messages are detected by RT-PCR (Parmantier et al., 1995). This could be

addressed by the isolation of RNA from purified DRG neurons which still show

expression of the -4/0kb lacZ transgene - at least in mixed cultures of dissociated DRG.

With the sensitive PMP22 TaqMan PCR a determination of the probably low expression

levels might be possible, although D'Urso et al., (1997) failed to detect PMP22 by

conventional RT-PCR in purified rat DRG neurons.

It is also possible that the -3/0 kb lacZ construct, which does not show any specific

expression as transgene, is silenced in vivo, for example by methylation. The previously


83

reported fact that many conserved CpG islands can be found around promoter 2 (van de

Wetering et al., 1999) and in front of exon 2 is consistent with such a possibility, since

methylation was proposed to be one regulatory mechanisms of promoter 2. In addition,

with the presence of CpG islands and the absence of a CCAAT box, promoter 2 shows a

structure typically found in promoters of house-keeping genes (van de Wetering et al.,

1999). It may therefore be that the -3/0kb Pro2 lacZ construct does not contain the

elements necessary to prevent silencing of the transgenic promoter 2.

Up to now, we focused our attention on the PNS expression of PMP22. But promoter 2

of PMP22 is active also in many non-neural tissues. Due to the broad PMP22 expression

pattern and due to the promoter structure as described above, promoter 2 was proposed to

be a constitutively active promoter (Suter et al., 1994, van de Wetering et al., 1999). So

far, I did not observe a consistent specific expression pattern of the transgenes analyzed

in the present in vivo promoter deletion study in non-neural tissues. This shows that the

transgenes did not contain the elements necessary for transgene specific expression in the

non-neural tissue in postnatal animals (data not shown). In combination with the results

from the -10/0kb PMP22 lacZ construct, which also did not show specific postnatal

expression in organs with high PMP22 mRNA levels such as small intestine and lung (cf.

Fig. 6-6, Suter et al., 1994), this indicates that I am missing regulatory elements

responsible for non-neural expression of PMP22 in the -10/0 kb region. As an inverse

interpretation I argued in the previous part (cf. part II), that this also could be due to the

presence of negative acting elements. In such a case, one would anticipate that a

dissection of the -10/0kb promoter region into subfragments would abrogate the effect of

negative acting elements in certain constructs. I did not observe such a phenomena and

therefore this interpretation is unlikely.

So far, no PMP22 promoter driven reporter construct completely reflected endogenous

PMP22 expression. The characterization of transgenic reporter constructs which contain

additional regions 5’ or 3’ to the previously characterized -10/0 kb genomic fragment,

will hopefully lead to the discovery of additional regulatory elements. Of special interest

would be those involved in the initiation of PMP22 expression in early development. The

detection of regions with high homology between mouse and human PMP22 genomic


84

sequences outside of the characterized -10/0 kb genomic fragment supports the

hypothesis that additional regulatory elements, for example between exon 2 and 3 or

downstream of exon 3 may contribute to PMP22 expression in early Schwann cells or in

non-neural tissues.

Our collaborators from the group of J. Snipes have shown that a -8.5/0kb PMP22 lacZ-

IRES-luciferase transgene with the rat promoter sequence shows only weak expression

in Schwann cells, but considerable expression in DRG (J. Snipes, personal

communication). The fact that many potential binding sites for Oct-6 and Brn-2 are

found in the -10/-8.5kb region and that the rat and mouse sequences are highly

conserved, make this element an attractive sequence to screen for binding of myelin

specific transcription factors. Further delineation of the 6kb LMSE with the -6.5/4kb

Pro1 lacZ and the -10/-6.5kb Pro1 lacZ construct is ongoing in our laboratory and will

hopefully answer the question of whether the LMSE can be taken apart further into

smaller segments. However, if it turns out that the -10/-6.5kb element leads to Schwann

cell expression, the indications from potential binding site would be confirmed and this -

10/-6.5kb or even the shorter -10/-8.5kb element indeed could be used to screen for

binding factors, for example with a yeast-one-hybrid system. Since this system was

successfully used only with short sequence elements (oligonucleotides)(Lehming et al.,

1994; Wang and Reed, 1993) the screening has to be performed only with subfragments

of the -10/-8.5kb element. If it is not possible to further narrow down the 6kb LSME,

potentially important regulatory regions on large sequence elements could be determined

by screening for DNaseI Hypersensitive sites (HSS). The challenge of these experiments

is to obtain enough starting material, since the most reasonable way to perform these

experiments is to take cell nuclei from myelinating Schwann cells. So far, at least the

isolation of nuclear extract from myelinating Schwann cells seems to be possible

(Forghani et al., 1999), and could be used for example also for electromobility shift

assays (EMSA). Classical EMSA permit the characterization of one or maybe a few

transcription factor at a time and for a DNA sequence usually between 20 bp and 300 bp.

Alternatively, if one can isolate sufficient amounts of nuclear extracts a new developed

system called “TranSignal Protein/DNA array” (Panomics, CA, USA) may allow

profiling the activities of multiple transcription factors simultaneously. With this system


85

one screens for the binding of transcription factors to a mixture of oligonucleotides with

known consensus-binding site sequences. Nuclear extracts are incubated with a mixture

of biotin-labelled oligonucleotides, DNA/protein complexes allowed to form, and DNA/

protein complexes are isolated. The recovered DNA is hybridized to a prespotted

membrane with complementary sequences to those of the oligonucleotides. However,

this system has the disadvantage that one is limited to known consensus binding

sequences. Further evaluation of this system will reveal how reliable and useful it is.

Alternatively, one may use chromatin immunoprecipitation assays combined with PCR

amplification of potential target regions to confirm a binding of a specific transcription

factor to a certain promoter region (Sasaki et al., 2002).

In summary, I have used a transgenic approach to characterize the sequences within the

PMP22 gene that are involved in its tissue- and cell-type specific as well as temporal

regulation. It will be interesting to further delineate the regions of the PMP22 gene and

its binding partners required for cell-type specificity and for regulation by neuron-

Schwann cell interactions. This will be supported by ongoing complementary in vitro

approaches, including the mapping of PMP22 promoters by transfection, by analyzing

DNA-proteins interactions (Hai et al., 2001; Saberan-Djoneidi et al., 2000), or by using

adenoviral infections of Schwann cells with Krox20 to study the regulatory network

involved in the initiation of PMP22 transcriptional expression. I anticipate that such

studies will provide additional insights into the coordinate regulation of myelin genes in

conjunction with related studies analyzing other genes encoding myelin components.

Furthermore, taking into account the strong upregulation of the -10/-4kb Pro1 lacZ

reporter gene during myelination and its Schwann cell-specificity, this promoter segment

is a valuable tool to specifically express potential myelin gene regulatory factors that

might also be involved in the pathogenesis of hereditary peripheral neuropathies, or to

regulate PMP22 expression levels by antisense approaches (Hai et al., 2001; Maycox et

al., 1997).

EXPERIMENTAL METHODS

87

9 EXPERIMENTAL METHODS

9.1 Generation of reporter constructs

For the generation of a construct expressing sh ble-lacZ under the control of PMP22

regulatory regions (-10/0kb PMP22 lacZ), a 10 kb Sal I/Not I fragment from cosmid

pTCF-6.1 (Magyar et al., 1996) was fused to a Not I-Nco I PCR fragment containing the

first 42 bp of the exon 2 and a Nco I-Sac II fragment of the sh ble-LacZ fusion gene

(Cayla, Toulouse, France) reconstituting the original Kozak sequence and translation

start on exon 2. The sequence of the PCR fragment was controlled by DNA sequencing.

The resulting 14 kb reporter construct (-10/0kb PMP22 lacZ; SNlacZ; SNEx2zeolacZ;

PMP22- lacZ) was excised from the vector backbone with Sal I and Sac II.

Four fragments were ligated for the generation of the -10/-4kb Pro1 lacZ (1AlacZ)

reporter construct: 1) a 3.5kb Sal I (-10kb) /Kpn I (-6.5kb) fragment derived from a Sal I

/Not I PMP22 subclone that was used to generate the -10/0kb PMP22 lacZ construct

(Figs. 5-1, 6-1, PMP22lacZ in Maier et al., 2002a) a 2.6kb Kpn I (-6.5kb)/Not I fragment

using a Not I site introduced by PCR at -3.91 kb with the same Sal I /Not I subclone as

template 3) a Not I (-0.12kb) /Bgl II fragment, containing exon 2 sequences and the sh

ble-LacZ fusion gene (pUT111, Cayla, Toulouse, France); and 4) a Sal I /Bgl II fragment

containing the Bluescript vector II KS+ (Stratagene). Fragments (3) and (4) were derived

from the PMP22 -10/0kb lacZ (PMP22lacZ) construct. The resulting reporter construct

was released from the vector backbone with Sal I and Sac II digestion for pronuclear

injection.

The -6.5/-4kb Pro1 lacZ (∆1AlacZ) was generated by an intramolecular deletion of a

3.5kb Kpn I(-10.1kb)/Kpn I(-6.5kb) fragment of the -10/-4kb Pro1 lacZ construct. To

generate the -10/-6.5kb Pro1 lacZ (∆Pro1lacZ) construct a Sal I (-10kb)/Kpn I (overhang

blunt-ended with T4 DNA Polymerase) fragment was fused to a Stu I(-4.33kb)/Sal I

fragment of the -10/-4kb Pro1 lacZ construct. The -10/-6.5kb Pro1 lacZ and the -6.5/-4kb

Pro1 lacZ constructs were released with a Kpn I and Sac II digest.

The -10/-4kb hsp lacZ (hspAlacZ) construct was derived from a fusion of three

fragments: a 3.5 kb Sal I /Kpn I fragment (the same as used for the -10/-4kb Pro1 lacZ

construct), a 2.2 kb Kpn I /Sal I fragment in which the Sal I site was introduced by PCR


88

at -4.29kb and a third Sal I /Sal I fragment containing the minimal 0.3kb hsp68 promoter

ligated to lacZ (clone p610ZA; R. Kothary, Universtity of Ottawa, Ottawa, Ontario,

Canada; (Forghani et al., 2001; Mandemakers et al., 2000) ). The sequence of all PCR

fragments was controlled by DNA sequencing. The resulting reporter construct was

excised from the vector backbone with Sma I and Sph I.

For the generation of the -3 kb Pro2 lacZ (PMP22-1B-lacZ; Maier et al., 2002) or -4 kb

Pro2 lacZ vector (1ABlacZ), a 7 kb respectively a 6 kb Sal I /Sac I fragment was

removed by Sal I and partial Sac I digest from the 5’ end of the -10/0kb PMP22 lacZ

(PMP22-lacZ) construct. The overhang was blunt-ended using T4 DNA polymerase and

the vector was intramolecularly ligated reconstituting the Sal I site. Both reporter

contructs were released with a Sal I and Sac II digest.

For the generation of the Del1AlacZ construct a Avr II(blunted)/Stu I deletion for

promoter 1 and Exon 1A was performed on a Kpn I/BamH I subclone of the -10/0 kb Sal

I/Not I fragment from cosmid pTCF-6.1 (Magyar et al., 1996) resulting in the plasmid

KBDel1A. The deletion was controlled by sequencing of the plasmid. In a next step a

3.0kb Kpn I/BamH I fragment of this KBDel1A plasmid was fused to a 6.3kb BamH I/

BamH I fragment, a 2.9kb BamH I/Sal I (containing the BS vector) and a 3.8 kb Sal I/

Kpn I fragment of the PMP22 -10/0kb lacZ construct.

For the generation of the Del1BlacZ construct a Bgl II/Avr II deletion for promoter 2 and

Exon 1B was performed on a BamH I/BamH I subclone (clone Bam2.3) of the PMP22

cosmid pTCF-6.1 (Magyar et al., 1996) resulting in the plasmid Bam2.3Del1B. The

deletion was controlled by sequencing of the plasmid. In a next step a BamH I/Not I

fragment of this Bam2.3Del1B plasmid was fused to a Not I/BamH I fragment

(containing BS vector and Exon1A) and a Not I/Not I fragment (containing the Ex2lacZ

cassette) of the PMP22 -10/0kb lacZ construct.

9.2 Generation of transgenic animals

The resulting reporter constructs were excised from the vector backbone as indicated,

purified by agarose gel electrophoresis, and electroeluted. The DNA was microinjected

into the pronucleus of fertilized eggs from B6D2F1 hybrid mice using standard


89

procedures. The founder animals were screened by PCR on DNA isolated from tail

biopsies with primers complementary to the lacZ sequence (lacZs: 5’-CCCATTACGG-

TCAATCCGCCG-3', LacZas: 5’-GCCTCCAGTACAGCGC-GGCTG-3’). For further

analysis the founders were mated with B6D2F1 hybrid mice. PCR positive founder

animals and PCR positive animals of the subsequent F1 generation were screened for the

expression of ß-gal in peripheral nerves on 50 µm cross-sections of tail biopsies (see next

section).

9.3 ß-gal histochemical analysis

Expression of the lacZ gene was monitored by standard histochemical staining reactions

with 5-bromo-4-chloro-3-indolyl-ß-galactopyranoside (X-gal; AppliChem, Germany). In

the spinal cord and brainstem analysis, mice were anesthetized and perfused

intracardially with 0.9% NaCl followed by 0.5% glutaraldehyde in 0.1M phosphate

buffer (pH 7.4). Tissues were dissected out and postfixed for 10 (sciatic nerve) to 60 min

at 4°C in 2% formaldehyde, 0.2% glutaraldehyde in phosphate buffered saline (PBS),

pH7.4. The specimens were washed three times with 1x PBS, equilibrated in 30%

sucrose overnight, and embedded in O.C.T. compound (Tissue Tek) for cryostat

sectioning. 30 µm-cryostat sections were stained for several hours (sciatic nerve) or

overnight (embryos, spinal cord sections) in the X-gal staining solution (5mM

K3[Fe(CN)6], 5mM K4[Fe(CN)6], 2mM MgCl2 , 1mg/ml X-Gal, in 1x PBS).

Embryos were fixed for 1 hour in 2% formaldehyde, 0.2% glutaraldehyde, 0.1% sodium

deoxycholate, 0.02% NP-40 in 1xPBS, washed three times in 1x PBS, cryoprotected

over night in 30% sucrose, and stored at –80°C. Whole mount stainings were performed

before cryoprotection in modified X-Gal staining solution (5mM 5mM K3[Fe(CN)6],

5mM K4[Fe(CN)6 ], 2mM MgCl2 , 0.1% sodium deoxycholate, 0.02% NP-40, 2mg/ml

X-Gal in 1x PBS). E19, P1 and P5 animals were decapitated and the head and two pieces

of the body were processed individually.

Sciatic nerves were dissected, fixed, teased in PBS buffer as described in (Neuberg et al.,

1999), transferred to glass slides, briefly air-dried, and incubated in the X-gal staining

solution.


90

Differences for part III:

Tissues were dissected out and fixed for 10 (sciatic nerve) to 45 min (spinal cord) in 2%

formaldehyde, 0.2% glutaraldehyde in phosphate buffered saline (PBS), pH7.4, at 4°C.

The specimens were washed three times with PBS and stained overnight in the X-gal

staining solution (5mM K3[Fe(CN)6], 5mM K4[Fe(CN)6], 2mM MgCl2 , 0.1% sodium

deoxycholate, 0.02% NP-40, 1mg/ml X-Gal, in 1x PBS) at 37°C. For X-gal staining of

teased sciatic nerve the nerves were fixed, the epineurium of the nerve was removed to

ensure penetration of the staining solution and stained overnight at 37°C. In a next step

the nerve was teased in PBS buffer as described in (Neuberg et al., 1999) and transferred

to glass slides and air-dried. Finally, teased fiber preparations were mounted in AF1

(Citifluor, Canterbury, UK) supplemented with DAPI (1:1000 Roche Diagnostics,

Switzerland). For the quantification of ß-gal positive internodes at least 100 DAPI-

stained nuclei corresponding to 100 internodes were counted for each animal and

analysed for associated X-Gal staining.

9.4 ß-gal solution assay

Organ samples (approx. 100mg) or 10-15mm of sciatic nerve from transgenic and

control mice were homogenized with a Polytron homogenizer in lysis buffer (0.1M

potassium phosphate buffer, pH 7.8; 0.2% Triton X-100, 0.5mM dithiothreitol). Cell

debris was removed by centrifugation and aliquots of each homogenate were stored at –

80°C. ß-gal activity was assayed in triplicates with the Galacto-StarTM kit (Tropix)

according to the manufacturers’ instructions. In brief, 10 µl of the lysate (or of a 1:50

dilution with lysis buffer in cases of very high ß-gal levels) were mixed with 100 µl of

reaction buffer. After 30-45 min the chemiluminescence was determined in a scintillation

counter (Canberra Packard SA). The amount of protein in each sample was measured

using the Bio-Rad DC protein assay in triplicate with BSA as a standard. The light

emission representing the enzymatic ß-gal activity (in relative light units RLU) was then

normalized to the amount of protein.

Differences for part III


91

For each time point 4 - 6 sciatic nerves from transgenic and control mice were

homogenized individually with a Polytron homogenizer, ß-gal activity was assayed in

triplicates with the Galacto-StarTM kit (Tropix) according to the manufacturers’

instructions and as described above.

9.5 Quantitative analysis of PMP22 mRNA levels

Total RNA was isolated using TRIzol reagent (Gibco BRL) according to the

manufacturer’s recommendations. Briefly, nerves were homogenized in the presence of

TRIzol with a Polytron homogenizer (PT 1200, Kinematica AG, Switzerland). Insoluble

material was removed from the homogenate by centrigugation for 10 min at 4°C, the

supernatant extracted with chloroform and precipitated with isopropanol. 500 ng of total

RNA was reverse transcribed in a 20 µl reaction using 140U MMLV reverse transcriptase

(Promega), 2.5 µM random hexamer primers, 1 mM dNTP, 30U RNasin in 1x Promega

reverse transcription buffer.

Five different PCR reactions were performed with the TaqMan PCR system to analyze

five different transcripts (Fig. 5-1): The 1A-PMP22 transcript with the forward primers

Pr1A (5’-GAGGAAGGGGTTACACCATTG-3’) located on exon 1A and the PMP22

sequence specific backward primer Pr2PMP22 (5’-GCAACACTAGC-ACCGCGAT-3’)

located in the second half of exon 2, the 1B-PMP22 message with Pr1B (5’-

TGTGCCTGAGGCTAATCTGC-3’) located in exon 1B and the primer Pr2PMP22, the

1A-lacZ transcript with Pr1A and the Pr2lacZ (5’-GTGAGCACCGGAACGGC-3’)

which is specific for the sh ble-lacZ message, and the 1B-lacZ transcript with primers

Pr1B and Pr2lacZ. To standardize between different samples, GAPDH mRNA levels

were analyzed with the primers GAPDH-f (5’-TGTGTCCGTCGTGGATCTGA-3’) and

GAPDH-b (5’-CCTGCTTCACC-ACCTTCTTGA-3’). To measure in real time the

amount of amplified PCR product, the TaqMan system (AP Applied Biosystems) was

used with the following sequence specific probes: 5’-

TCCTCTGATCCCGAGCCCAACTCC-3’ (with 5’ FAM reporter and 3’ TAMRA

quencher dye modifications) located in the first half of exon 2 in the common sequence

for the 1A-PMP22, 1B-PMP22, 1A-lacZ and 1B-lacZ message, and 5’-


92

CCGCCTGGAGAAACCTGCCAAGTATG-3’ (5’ VIC, 3’ TAMRA) specific for the

GAPDH message.

Purified PCR products were prepared as serial dilutions from 10-9 to 10-15 M for the

construction of the standard curves. Amplifications were performed in triplicate 25 µl

reactions for 40 cycles (denaturation at 95°C for 20s, annealing at 53°C for 30s,

elongation at 60°C for 60s) on the ABI Prism 7700 sequence detection system. The PCR

reaction contained 1mM MgCl2, 2.8% DMSO, 500nM forward and backward primers,

200nM TaqMan probe and 1 µl template from the RT reaction in addition to 12.5 µl 2x

TaqMan Universal PCR Master Mix (AP Applied Biosystems). The quantitation of gene

expression was performed as described in the User Bulletin #2, ABI PRISM 7700

Sequence Detection System, PE Applied Biosystems.

9.6 -10/0kb PMP22 lacZ x PMP22-/- and -10/0kb PMP22lacZ x Tr mice

To obtain -10/0kb PMP22 lacZ transgenic mice on a PMP22-deficient background (-10/

0kb PMP22 lacZ; PMP22-/-) in a first breeding -10/0kb PMP22 lacZ transgenic animals

Line 48.4 were mated with PMP22-/- females (Adlkofer et al., 1995). In a second

breeding -10/0kb PMP22 lacZ transgenic mice heterozygous for PMP22 (-10/0kb

PMP22 lacZ; PMP22 wt/-) were mated with PMP22-/- females. Homogenates of single

sciatic nerves of -10/0kb PMP22 lacZ / PMP22-/- animals were prepared as described

and compared with age-matched -10/0kb PMP22 lacZ transgenic animals without

PMP22 mutations. PCR analysis of PMP22-deficient mice has been reported elsewhere

(Sancho et al., 2001).

-10/0kb PMP22 lacZ transgenic animals with the Tr point mutation in the PMP22 gene

(Suter et al., 1992b) were obtained by mating -10/0kb PMP22 lacZ transgenic mice Line

48.4 with females heterozygous for the Tr mutation (Tr/+). ß-gal solution assays were

performed if possible with siblings otherwise with age-matched animals with or without

the PMP22 Trembler mutation at postnatal day 21 and in adult P60 and P90 animals.

PCR analysis of Tr mice has been reported elsewhere (Adlkofer et al., 1997b).


93

9.7 Sciatic nerve transsection and crush

Using aseptic technique, the sciatic nerve of anesthetized (Ketamine 80mg/kg and

Xylazine 4mg/kg) adult (9-12 weeks old) -10/0kb PMP22 lacZ transgenic mice were

exposed at the sciatic notch. Nerves were doubly ligated, transsected with fine scissors

and the nerve-stumps were sutured at least five millimeter apart to prevent regeneration.

Nerve crush was produced by tightly compressing the sciatic nerve at the sciatic notch

with flattened forceps twice, each time for 10s; this technique causes the axons to

degenerate, but allows axonal regeneration. At varying times after nerve injury, 3-4

animals for each time point were sacrified, the distal nerve-stumps were removed, the

most proximal 4-5 mm trimmed off and further processed as described earlier. For

transsected nerves, the entire distal nerve-stump was taken just below the lesion. At each

time point the homogenate of the distal part of the lesioned nerve was compared with the

equivalent part of the contralateral unlesioned nerve.

9.8 Immunocytochemistry of dissociated dorsal root ganglia(DRG)

Dorsal root ganglia of E19 embryos were isolated, digested with 0.2% trypsin for 20

min, triturated after the addition of DMEM/10%FCS, washed once with DMEM/

10%FCS, and plated on poly-L-lysin coated plastic dishes (Corning) in 10%FCS/

DMEM/NGF (50ng/ml, Sigma). After 3 days, cells were washed, fixed for 10 min with

4% formaldehyde, incubated for 4-6 hours with blocking solution (10% NGS, 0.1%

Triton X-100, 1% BSA in 1x PBS) and stained with rabbit polyclonal antibodies against

ß-galactosidase (1:300, CN Kappel) for 12-14 hours and with mouse monoclonal

antibody against NF160 (1:500, Sigma) for 1 hour, washed with PBS and incubated for 1

hour with the secondary antibodies goat anti-mouse Cy3 and goat anti-rabbit FITC

(1:300, Jackson Laboratories). Immunoreactivity was visualized by conventional

fluorescence microscopy using a Hamamatsu Colour Chilled 3CCD Camera in

conjunction with Adobe Photoshop 5.0.


94

Differences for part III

Dorsal root ganglia of postnatal day P4 mice were isolated and digested with 0.25%

trypsin and 0.3mg/ml collagenase type I (Worthington) in Ca2+/Mg2+ free Hank’s

Balanced Salt Solution for 45 min at 37°C. After the addition of 100µl FCS/ml to the

digestion mix, cells were triturated and further processed as discribed above.

9.9 Cell culture, transfection and reporter assays

Promoter deletion study in MSC80 and NIH3T3 cells

The mouse Schwann cell line MSC80 (Boutry et al., 1992) and the NIH 3T3 cell line

were transfected using SuperFect reagent (Qiagen, Switzerland) according to the

manufacturer’s recommendations. Equimolar amounts of -10/0kb PMP22 lacZ, -3/0 kb

Pro1 lacZ, -4/0 kb Pro1 lacZ, Del1AlacZ, Del1BlacZ, pSV-ß-galatosidase control vector

(Promega) or empty lacZ vector (pUT111, Cayla, Toulouse, France) were used as

reporter constructs and total amount of plasmids was kept constant by addition of empty

Bluescript Vector (Stratagene). The plasmid SV40Luciferase (pGL3-Promoter Vector,

Promega) was used as an internal control to assess transfection efficiency and for

normalization. Cells were maintained in DMEM supplemented with 10% FCS and

10µM forskolin if indicated. Fourty hours after transfection the cells were washed once

with 1x PBS, lysed in 1x reporter lysis buffer (Promega) and extracts were assayed for ß-

gal activity in microtiter plates in triplicate according to (Sambrook et al., 1989) and for

luciferase activity with the LucLite Plus Luminescence kit (Packard Biosciences).

Incucible Sox10 expression in N2A cells

RNA was extracted with the Trizol Method from N2A cells stably transfected with the

reverse tetracycline-controlled transactivator (rtTA, harboring the G418 resistance) and

the cDNA of Sox10 under an tetracycline-regulatable promoters (harboring a

hygromycin selection cassette). The cells were maintained in DMEM + 0.5% FCS and

Sox10 expression was induced for 48 hours with 5µg/ml Doxycylin. The generation of

this N2A cell line, the PCR Primers and cDNA Synthesis is described in (Peirano et al.,


95

2000). To standardize between different samples, 18S rRNA levels were measured with a

predevelopped assay reagent from Applied Biosystems.

Screening for upregulated transcription factors with quantitative PCR

Subconfluent (40-60%) rat Schwann cells were infected overnight with the Adenovirus

Adegr2GFP and AdGFPSJS2 or AdGFPR1* (Ehrengruber et al., 2000) as control virus

either in DMEM Medium in the presence of 10% FCS or in defined N2 medium after 48

hours. Total RNA was isolated from the cells after 30, 24 or 48 hours as indicated with

the RNeasy Total RNA Purification Mini Prep Kit (Qiagen, Switzerland) according to the

manufacturers’ instructions. Homogenisation of cell lysates was done using the syringe

method by passing the lysate three to four times through a sterile plastic syringe fitted

with a 18-21 gauche needle. Reverse transcription was performed as described in section

9.5 above. SYBR Green PCR were performed at standard conditions as recommended by

Applied Biosystems (25mM MgCl2, 12mM dNTP (containing dUTP), Amp Erase,

AmpliTaq Gold, 500nM forward and backward primer, 3µl of 1:5 diluted RT reaction).

To standardize between different samples, 18S rRNA levels were measured with a

predevelopped assay reagent from Applied Biosystems with TaqMan PCR. The

Oligonucletides used in the quantitative SYBR Green PCRs are:

Oligo# Sequence (5’-3’) Primer/PCR Name

59 CACGGGCCAGGAGCG mrBrn2.1-f60 TTGGCAGCGTGGTGCAC mrBrn2.1-b61 GGCCTTGGGCATACTCAACA mrStAR-f62 CGTCCCCGTTCTCCTGC mStAR-b81 CACCTTACTTAGCACTTCATCTCCAT rSTAR-b63 CTGGTCAGCGTGACAAAGTACAG mKZF1-f64 TTATCACAATTGAAAAGCTTCTTTGG mKZF1-b65 GGGATCCTGTTCCTGCACAT m(r)totPMP-for66 TGCCAGAGATCAGTCGTGTGT m(r)tg+totPMP-b67 TCCACCATCGTCAGCCAATGGCT mr PMP22TqMprobe72 GGCCTCACACCCGCC Pro1C9973Tq-f73 GCCTTACCTGTCCCAGTTAGGG mrIREBP1-f74 ACTGTTGCCGATGCAGGTC mrIREBP1-b75 ACGAGCTGCACGGGCC mrBrn2.2nd-f76 TTGGCAGCGTGGTGCAC mrBrn2.2nd-b79 AAATTGCATGGAAAAGGGCA mTRIP8-f80 GTGGCCTCGCACGCA mTRIP8-b82 TCCTTACTCTTCAGTCAGCAAATAGCTA rKZF1-f83 TGGGCTATGTGATAATGGATCAAT rKZF1-b


96

84 ACAACGAGCTCTCTCACTTCCTG mrS100ß-f85 CGTCCAGCGTCTCCATCAC mrS100ß-b90 CATGGGCAAATTCTCCATTGA mrEgr2-f91 TTGCAAGATGCCCGCAC mrEgr2-b92 AAGGAATCTTTGTCCGCGAG periaxin (L+S)_2 f93 CTCAAAGAACACACGGGCG periaxin (L+S)_2 r94 CTGGCAGAGCGGATAACAC rNAB2-f95 GTGTTCGAGGCAGTGCCAT rNAB2-b99 GGGAGCCGAGCGAACAA rEGR1-f101 CACCAGCGCCTTCTCGTTATCCC rEGR1-b102 TCTGTACCCCGAGGAGATCC rEGR3-f104 TCCATCACATTCTCTGTAGCCATC rEGR3-b108 CCCTGGCCATTGTGGTTTAC mrMPZ-f109 CCATTCACTGGACCAGAAGGAG mrMPZ-b

9.10 Sequence analysis and determination of potentialbinding sites

Global alignments of orthologous PMP22 loci were done with available murine

(accession number AL 592215.12) and human (AC 005703) sequences containing the

PMP22 gene loci, using VISTA homology plot software (http://www-gsd.lbl.gov/vista/

index.html; Dubchak et al., 2000; Mayor et al., 2000). The results obtained were

compared with the percent identity plot (PIP) constructed with Pipmaker software at

http://bio.cse.psu.edu/pipmaker (data not shown). The three species comparison of the -

10/0kb region was done with VISTA homology plot software with orthologous

sequences of mouse, human, and rat (AC 108967).

To screen the murine genomic DNA sequences for interspersed repeats and low

complexity DNA sequences, the RepeatMasker program was used (Smit, AFA & Green,

P RepeatMasker at http://ftp.genome.washington.edu/RM/RepeatMasker.html and

references therein).

Screening for potential binding site was performed on the mouse -10/0kb fragment (AL

592215.12) with the MatInspector software (Quandt et al., 1995) with all available

vertebrate matrices (core similarity: 0.75; matrix similarity: optimized; Matrix Family

Library Version 2.4 May 2002). The software is availabe online at http://

www.genomatix.de and is based on the TRANSFAC databases (Heinemeyer et al., 1998)

(http://transfac.gbf.de/).

REFERENCES

97

10 REFERENCES

Adlkofer, K., Frei, R., Neuberg, D. H., Zielasek, J., Toyka, K. V., and Suter, U. (1997a).Heterozygous peripheral myelin protein 22-deficient mice are affected by a progressivedemyelinating tomaculous neuropathy. J Neurosci 17: 4662-71.

Adlkofer, K., Martini, R., Aguzzi, A., Zielasek, J., Toyka, K. V., and Suter, U. (1995).Hypermyelination and demyelinating peripheral neuropathy in Pmp22- deficient mice.Nat. Genet. 11: 274-280.

Adlkofer, K., Naef, R., and Suter, U. (1997b). Analysis of compound heterozygous mice revealsthat the Trembler mutation can behave as a gain-of-function allele. J Neurosci Res 49:671-80.

Aguayo, A. J., Charron, L., and Bray, G. M. (1976). Potential of Schwann cells fromunmyelinated nerves to produce myelin: a quantitative ultrastructural and radiographicstudy. J Neurocytol 5: 565-73.

Aguayo, A. J., Epps, J., Charron, L., and Bray, G. M. (1976). Multipotentiality of Schwann cellsin cross-anastomosed and grafted myelinated and unmyelinated nerves: quantitativemicroscopy and radioautography. Brain Res 104: 1-20.

Amato, A. A., Gronseth, G. S., Callerame, K. J., Kagan-Hallet, K. S., Bryan, W. W., and Barohn,R. J. (1996). Tomaculous neuropathy: a clinical and electrophysiological study in patientswith and without 1.5-Mb deletions in chromosome 17p11.2. Muscle Nerve 19: 16-22.

Anderson, T. J., Klugmann, M., Thomson, C. E., Schneider, A., Readhead, C., Nave, K. A., andGriffiths, I. R. (1999). Distinct phenotypes associated with increasing dosage of the PLPgene: implications for CMT1A due to PMP22 gene duplication. Ann N Y Acad Sci 883:234-46.

Aoyama, A., Tamura, T. A., and Mikoshiba, K. (1990). Regulation of brain-specific transcriptionof the mouse myelin basic protein gene: function of the NFI-binding site in the distalpromoter. Biochem Biophys Res Commun 167: 648-53.

Araki, T., Nagarajan, R., and Milbrandt, J. (2001). Identification of genes induced in peripheralnerve after injury. Expression profiling and novel gene discovery. J Biol Chem 276:34131-41.

Awatramani, R., Shumas, S., Kamholz, J., and Scherer, S. S. (2002). TGFbeta1 modulates thephenotype of Schwann cells at the transcriptional level. Mol Cell Neurosci 19: 307-19.

Baechner, D., Liehr, T., Hameister, H., Altenberger, H., Grehl, H., Suter, U., and Rautenstrauss,B. (1995). Widespread expression of the peripheral myelin protein-22 gene (PMP22) inneural and non-neural tissues during murine development. J Neurosci Res 42: 733-41.

Beckmann, A. M., and Wilce, P. A. (1997). Egr transcription factors in the nervous system.Neurochem Int 31: 477-510; discussion 517-6.

Bellefroid, E. J., Sahin, M., Poncelet, D. A., Riviere, M., Bourguignon, C., Martial, J. A., Morris,P. L., Pieler, T., Szpirer, C., and Ward, D. C. (1998). Kzf1 - a novel KRAB zinc fingerprotein encoding gene expressed during rat spermatogenesis. Biochim Biophys Acta 1398:321-9.

REFERENCES

98

Benbrook, D. M., and Jones, N. C. (1994). Different binding specificities and transactivation ofvariant CRE's by CREB complexes. Nucleic Acids Res 22: 1463-9.

Berger, P., Young, P., and Suter, U. (2002). Molecular cell biology of Charcot-Marie-Toothdisease. Neurogenetics 4: 1-15.

Bermingham, J. R., Jr., Scherer, S. S., O'Connell, S., Arroyo, E., Kalla, K. A., Powell, F. L., andRosenfeld, M. G. (1996). Tst-1/Oct-6/SCIP regulates a unique step in peripheralmyelination and is required for normal respiration. Genes Dev 10: 1751-62.

Berndt, J. A., Kim, J. G., and Hudson, L. D. (1992). Identification of cis-regulatory elements inthe myelin proteolipid protein (PLP) gene. J Biol Chem 267: 14730-7.

Bharucha, V. A., Peden, K. W., Subach, B. R., Narayanan, V., and Tennekoon, G. I. (1993).Characterization of the cis-acting elements of the mouse myelin P2 promoter. J NeurosciRes 36: 508-19.

Bharucha, V. A., Peden, K. W., and Tennekoon, G. I. (1994). SV40 large T antigen with c-Jundown-regulates myelin P0 gene expression: a mechanism for papovaviral T antigen-mediated demyelination. Neuron 12: 627-37.

Boerkoel, C. F., Takashima, H., Bacino, C. A., Daentl, D., and Lupski, J. R. (2001). EGR2mutation R359W causes a spectrum of Dejerine-Sottas neuropathy. Neurogenetics 3: 153-7.

Borycki, A. G., and Emerson, C. P. (1997). Muscle determination: another key player inmyogenesis? Curr Biol 7: R620-3.

Bosse, F., Brodbeck, J., and Muller, H. W. (1999). Post-transcriptional regulation of theperipheral myelin protein gene PMP22/gas3. J Neurosci Res 55: 164-77.

Boutry, J. M., Hauw, J. J., Gansmuller, A., Di-Bert, N., Pouchelet, M., and Baron-VanEvercooren, A. (1992). Establishment and characterization of a mouse Schwann cell linewhich produces myelin in vivo. J Neurosci Res 32: 15-26.

Brancolini, C., Marzinotto, S., Edomi, P., Agostoni, E., Fiorentini, C., Muller, H. W., andSchneider, C. (1999). Rho-dependent regulation of cell spreading by the tetraspanmembrane protein Gas3/PMP22. Mol Biol Cell 10: 2441-59.

Brancolini, C., Edomi, P., Marzinotto, S., and Schneider, C. (2000). Exposure at the cell surfaceis required for gas3/PMP22 To regulate both cell death and cell spreading: implication forthe Charcot-Marie- Tooth type 1A and Dejerine-Sottas diseases. Mol Biol Cell 11: 2901-14.

Britsch, S., Goerich, D. E., Riethmacher, D., Peirano, R. I., Rossner, M., Nave, K. A.,Birchmeier, C., and Wegner, M. (2001). The transcription factor Sox10 is a key regulatorof peripheral glial development. Genes Dev 15: 66-78.

Brown, A. M., and Lemke, G. (1997). Multiple regulatory elements control transcription of theperipheral myelin protein zero gene. J Biol Chem 272: 28939-47.

Campagnoni, A. T., Pribyl, T. M., Campagnoni, C. W., Kampf, K., Amur-Umarjee, S., Landry, C.F., Handley, V. W., Newman, S. L., Garbay, B., and Kitamura, K. (1993). Structure anddevelopmental regulation of Golli-mbp, a 105-kilobase gene that encompasses the myelin

REFERENCES

99

basic protein gene and is expressed in cells in the oligodendrocyte lineage in the brain. JBiol Chem 268: 4930-8.

Chalepakis, G., and Gruss, P. (1995). Identification of DNA recognition sequences for the Pax3paired domain. Gene 162: 267-70.

Chan, J. R., Rodriguez-Waitkus, P. M., Ng, B. K., Liang, P., and Glaser, M. (2000). Progesteronesynthesized by Schwann cells during myelin formation regulates neuronal geneexpression. Mol Biol Cell 11: 2283-95.

Chance, P. F., Alderson, M. K., Leppig, K. A., Lensch, M. W., Matsunami, N., Smith, B.,Swanson, P. D., Odelberg, S. J., Disteche, C. M., and Bird, T. D. (1993). DNA deletionassociated with hereditary neuropathy with liability to pressure palsies. Cell 72: 143-151.

D'Urso, D., Schmalenbach, C., Zoidl, G., Prior, R., and Muller, H. W. (1997). Studies on theeffects of altered PMP22 expression during myelination in vitro. J Neurosci Res 48: 31-42.

De Leon, M., Nahin, R. L., Mendoza, M. E., and Ruda, M. A. (1994). SR13/PMP-22 expressionin rat nervous system, in PC12 cells, and C6 glial cell lines. J Neurosci Res 38: 167-81.

Desarnaud, F., Do Thi, A. N., Brown, A. M., Lemke, G., Suter, U., Baulieu, E. E., andSchumacher, M. (1998). Progesterone stimulates the activity of the promoters ofperipheral myelin protein-22 and protein zero genes in Schwann cells. J Neurochem 71:1765-8.

Devine-Beach, K., Lashgari, M. S., and Khalili, K. (1990). Myelin basic protein genetranscription. Identification of proximal and distal cis-acting regulatory elements. J BiolChem 265: 13830-5.

Dickson, K. M., Bergeron, J. J., Shames, I., Colby, J., Nguyen, D. T., Chevet, E., Thomas, D. Y.,and Snipes, G. J. (2002). Association of calnexin with mutant peripheral myelin protein-22ex vivo: a basis for "gain-of-function" ER diseases. Proc Natl Acad Sci U S A 99: 9852-7.

Donahue, L. M., and Reinhart, A. J. (1998). POU domain genes are differentially expressed inthe early stages after lineage commitment of the PNS-derived stem cell line, RT4-AC.Brain Res Dev Brain Res 106: 1-12.

Drocourt, D., Calmels, T., Reynes, J. P., Baron, M., and Tiraby, G. (1990). Cassettes of theStreptoalloteichus hindustanus ble gene for transformation of lower and higher eukaryotesto phleomycin resistance. Nucleic Acids Res 18: 4009.

Dubchak, I., Brudno, M., Loots, G. G., Pachter, L., Mayor, C., Rubin, E. M., and Frazer, K. A.(2000). Active conservation of noncoding sequences revealed by three-way speciescomparisons. Genome Res 10: 1304-6.

Duchala, C. S., Fuss, B., and Macklin, W. B. (1996). Use of Transgenic Systems to InvestigateOligodendrocyte Differentiation and Function. Methods 10: 478-88.

Dyck, P. J., Chance, P., Lebo, R., and Carney, J. A. (1993). Hereditary motor and sensoryneuropathies. In: Dyck PJ, Thomas PK, Grffin JW, Low PA, Poduslo JF, editors.Peripheral Neuropathy Philadelphia: Saunders, 1993:: 1094-1136.

Ehrengruber, M. U., Muhlebach, S. G., Sohrman, S., Leutenegger, C. M., Lester, H. A., andDavidson, N. (2000). Modulation of early growth response (EGR) transcription factor-dependent gene expression by using recombinant adenovirus. Gene 258: 63-9.

REFERENCES

100

Elliott, J. I., Festenstein, R., Tolaini, M., and Kioussis, D. (1995). Random activation of atransgene under the control of a hybrid hCD2 locus control region/Ig enhancer regulatoryelement. Embo J 14: 575-84.

Fabbretti, E., Edomi, P., Brancolini, C., and Schneider, C. (1995). Apoptotic phenotype inducedby overexpression of wild-type gas3/PMP22: its relation to the demyelinating peripheralneuropathy CMT1A. Genes Dev 9: 1846-1856.

Farsetti, A., Desvergne, B., Hallenbeck, P., Robbins, J., and Nikodem, V. M. (1992).Characterization of myelin basic protein thyroid hormone response element and itsfunction in the context of native and heterologous promoter. J Biol Chem 267: 15784-8.

Farsetti, A., Mitsuhashi, T., Desvergne, B., Robbins, J., and Nikodem, V. M. (1991). Molecularbasis of thyroid hormone regulation of myelin basic protein gene expression in rodentbrain. J Biol Chem 266: 23226-32.

Feltri, M. L., D'Antonio, M., Quattrini, A., Numerato, R., Arona, M., Previtali, S., Chiu, S. Y.,Messing, A., and Wrabetz, L. (1999). A novel P0 glycoprotein transgene activatesexpression of lacZ in myelin-forming Schwann cells. Eur J Neurosci 11: 1577-86.

Fiering, S., Whitelaw, E., and Martin, D. I. (2000). To be or not to be active: the stochastic natureof enhancer action. Bioessays 22: 381-7.

Forghani, R., Garofalo, L., Foran, D. R., Farhadi, H. F., Lepage, P., Hudson, T. J., Tretjakoff, I.,Valera, P., and Peterson, A. (2001). A distal upstream enhancer from the myelin basicprotein gene regulates expression in myelin-forming schwann cells. J Neurosci 21: 3780-7.

Forghani, R., Nesbitt, J., Snipes, J., Shooter, E. M., and Peterson, A. (1999). Preparation ofnuclear extracts from myelinating Schwann cells. J Neurosci Methods 89: 129-32.

Friede, R. L., and Samorajski, T. (1968). Myelin formation in the sciatic nerve of the rat. Aquantitative electron microscopic, histochemical and radioautographic study. JNeuropathol Exp Neurol 27: 546-70.

Garbay, B., Boiron-Sargueil, F., and Cassagne, C. (1995). Expression of the exon 1A-containingPMP22 transcript is altered in the trembler mouse. Neurosci Lett 198: 157-60.

Garbern, J. Y., Cambi, F., Tang, X. M., Sima, A. A., Vallat, J. M., Bosch, E. P., Lewis, R., Shy, M.,Sohi, J., Kraft, G., Chen, K. L., Joshi, I., Leonard, D. G., Johnson, W., Raskind, W.,Dlouhy, S. R., Pratt, V., Hodes, M. E., Bird, T., and Kamholz, J. (1997). Proteolipid proteinis necessary in peripheral as well as central myelin. Neuron 19: 205-18.

Gatignol, A., Durand, H., and Tiraby, G. (1988). Bleomycin resistance conferred by a drug-binding protein. FEBS Lett 230: 171-5.

Ghazvini, M., Mandemakers, W., Jaegle, M., Piirsoo, M., Driegen, S., Koutsourakis, M., Smit,X., Grosveld, F., and Meijer, D. (2002). A cell type-specific allele of the POU gene Oct-6reveals Schwann cell autonomous function in nerve development and regeneration. EmboJ 21: 4612-4620.

Ghislain, J., Desmarquet-Trin-Dinh, C., Jaegle, M., Meijer, D., Charnay, P., and Frain, M. (2002).Characterisation of cis-acting sequences reveals a biphasic, axon- dependent regulation ofKrox20 during Schwann cell development. Development 129: 155-66.

REFERENCES

101

Goujet-Zalc, C., Babinet, C., Monge, M., Timsit, S., Cabon, F., Gansmuller, A., Miura, M.,Sanchez, M., Pournin, S., Mikoshiba, K., and et al. (1993). The proximal region of theMBP gene promoter is sufficient to induce oligodendroglial-specific expression intransgenic mice. Eur J Neurosci 5: 624-32.

Graubert, T. A., Hug, B. A., Wesselschmidt, R., Hsieh, C. L., Ryan, T. M., Townes, T. M., andLey, T. J. (1998). Stochastic, stage-specific mechanisms account for the variegation of ahuman globin transgene. Nucleic Acids Res 26: 2849-58.

Gravel, M., Di Polo, A., Valera, P. B., and Braun, P. E. (1998). Four-kilobase sequence of themouse CNP gene directs spatial and temporal expression of lacZ in transgenic mice. JNeurosci Res 53: 393-404.

Grubinska, B., Laszkiewicz, I., Royland, J., Wiggins, R. C., and Konat, G. W. (1994).Differentiation-specific demethylation of myelin associated glycoprotein gene in culturedoligodendrocytes. J Neurosci Res 39: 233-42.

Guillot, P. V., Liu, L., Kuivenhoven, J. A., Guan, J., Rosenberg, R. D., and Aird, W. C. (2000).Targeting of human eNOS promoter to the Hprt locus of mice leads to tissue-restrictedtransgene expression. Physiol Genomics 2: 77-83.

Gupta, S. K., Pringle, J., Poduslo, J. F., and Mezei, C. (1993). Induction of myelin genes duringperipheral nerve remyelination requires a continuous signal from the ingrowing axon. JNeurosci Res 34: 14-23.

Hagedorn, L., Suter, U., and Sommer, L. (1999). P0 and PMP22 mark a multipotent neural crest-derived cell type that displays community effects in response to TGF-beta family factors.Development 126: 3781-94.

Hai, M., Bidichandani, S. I., Hogan, M. E., and Patel, P. I. (2001). Competitive binding oftriplex-forming oligonucleotides in the two alternate promoters of the PMP22 gene.Antisense Nucleic Acid Drug Dev 11: 233-46.

Hai, M., Bidichandani, S. I., and Patel, P. I. (2001). Identification of a positive regulatory elementin the myelin-specific promoter of the PMP22 gene. J Neurosci Res 65: 508-19.

Haney, C., Snipes, G. J., Shooter, E. M., Suter, U., Garcia, C., Griffin, J. W., and Trapp, B. D.(1996). Ultrastructural distribution of PMP22 in Charcot-Marie-Tooth disease type 1A. JNeuropathol Exp Neurol 55: 290-9.

Hardison, R., Slightom, J. L., Gumucio, D. L., Goodman, M., Stojanovic, N., and Miller, W.(1997). Locus control regions of mammalian beta-globin gene clusters: combiningphylogenetic analyses and experimental results to gain functional insights. Gene 205: 73-94.

He, X., Gerrero, R., Simmons, D. M., Park, R. E., Lin, C. J., Swanson, L. W., and Rosenfeld, M.G. (1991). Tst-1, a member of the POU domain gene family, binds the promoter of thegene encoding the cell surface adhesion molecule P0. Mol Cell Biol 11: 1739-44.

He, X., Treacy, M. N., Simmons, D. M., Ingraham, H. A., Swanson, L. W., and Rosenfeld, M. G.(1989). Expression of a large family of POU-domain regulatory genes in mammalian braindevelopment. Nature 340: 35-41.

Heinemeyer, T., Wingender, E., Reuter, I., Hermjakob, H., Kel, A. E., Kel, O. V., Ignatieva, E. V.,Ananko, E. A., Podkolodnaya, O. A., Kolpakov, F. A., Podkolodny, N. L., and Kolchanov,

REFERENCES

102

N. A. (1998). Databases on transcriptional regulation: TRANSFAC, TRRD and COMPEL.Nucleic Acids Res 26: 362-7.

Huehne, K., and Rautenstrauss, B. (2001). Transcriptional startpoints and methylation patterns inthe PMP22 promoters of peripheral nerve, leukocytes and tumor cell lines. Int J Mol Med7: 669-75.

Huhne, K., Park, O., Liehr, T., and Rautenstrauss, B. (1999). Expression analysis of the PMP22gene in glioma and osteogenic sarcoma cell lines. J Neurosci Res 58: 624-31.

Hume, D. A. (2000). Probability in transcriptional regulation and its implications for leukocytedifferentiation and inducible gene expression. Blood 96: 2323-8.

Huxley, C., Passage, E., Manson, A., Putzu, G., Figarella-Branger, D., Pellissier, J. F., andFontes, M. (1996). Construction of a mouse model of Charcot-Marie-Tooth disease type1A by pronuclear injection of human YAC DNA. Hum Mol Genet 5: 563-9.

Ikenaka, K., and Kagawa, T. (1995). Transgenic systems in studying myelin gene expression.Dev Neurosci 17: 127-36.

Imada, M., and Sueoka, N. (1978). Clonal sublines of rat neurotumor RT4 and celldifferentiation. I. Isolation and characterization of cell lines and cell type conversion. DevBiol 66: 97-108.

Inoue, K., Dewar, K., Katsanis, N., Reiter, L. T., Lander, E. S., Devon, K. L., Wyman, D. W.,Lupski, J. R., and Birren, B. (2001). The 1.4-Mb CMT1A duplication/HNPP deletiongenomic region reveals unique genome architectural features and provides insights into therecent evolution of new genes. Genome Res 11: 1018-33.

Jaegle, M., Mandemakers, W., Broos, L., Zwart, R., Karis, A., Visser, P., Grosveld, F., and Meijer,D. (1996). The POU factor Oct-6 and Schwann cell differentiation. Science 273: 507-10.

Jaegle, M., and Meijer, D. (1998). Role of Oct-6 in Schwann cell differentiation. Microsc ResTech 41: 372-8.

Jessen, K. R., and Mirsky, R. (1999). Schwann cells and their precursors emerge as majorregulators of nerve development. Trends Neurosci 22: 402-10.

Kagawa, T., Ikenaka, K., Inoue, Y., Kuriyama, S., Tsujii, T., Nakao, J., Nakajima, K., Aruga, J.,Okano, H., and Mikoshiba, K. (1994). Glial cell degeneration and hypomyelination causedby overexpression of myelin proteolipid protein gene. Neuron 13: 427-42.

Kaku, D. A., Parry, G. J., Malamut, R., Lupski, J. R., and Garcia, C. A. (1993). Nerve conductionstudies in Charcot-Marie-Tooth polyneuropathy associated with a segmental duplicationof chromosome 17. Neurology 43: 1806-8.

Kallunki, P., Edelman, G. M., and Jones, F. S. (1998). The neural restrictive silencer element canact as both a repressor and enhancer of L1 cell adhesion molecule gene expression duringpostnatal development. Proc Natl Acad Sci U S A 95: 3233-8.

Kallunki, P., Edelman, G. M., and Jones, F. S. (1997). Tissue-specific expression of the L1 celladhesion molecule is modulated by the neural restrictive silencer element. J Cell Biol 138:1343-54.

REFERENCES

103

Kamholz, J., Sessa, M., Scherer, S., Vogelbacker, H., Mokuno, K., Baron, P., Wrabetz, L., Shy,M., and Pleasure, D. (1992). Structure and expression of proteolipid protein in theperipheral nervous system. J Neurosci Res 31: 231-44.

Kothary, R., Clapoff, S., Brown, A., Campbell, R., Peterson, A., and Rossant, J. (1988). Atransgene containing lacZ inserted into the dystonia locus is expressed in neural tube.Nature 335: 435-7.

Kothary, R., Clapoff, S., Darling, S., Perry, M. D., Moran, L. A., and Rossant, J. (1989).Inducible expression of an hsp68-lacZ hybrid gene in transgenic mice. Development 105:707-14.

Kuhlbrodt, K., Herbarth, B., Sock, E., Hermans-Borgmeyer, I., and Wegner, M. (1998). Sox10, anovel transcriptional modulator in glial cells. J Neurosci 18: 237-50.

Kuhn, R., Monuki, E. S., and Lemke, G. (1991). The gene encoding the transcription factor SCIPhas features of an expressed retroposon. Mol Cell Biol 11: 4642-50.

Laszkiewicz, I., Grubinska, B., Wiggins, R. C., and Konat, G. W. (1997). Structuralcharacterization of myelin-associated glycoprotein gene core promoter. J Neurosci Res 50:928-36.

Lee, M., Brennan, A., Blanchard, A., Zoidl, G., Dong, Z., Tabernero, A., Zoidl, C., Dent, M. A.,Jessen, K. R., and Mirsky, R. (1997). P0 is constitutively expressed in the rat neural crestand embryonic nerves and is negatively and positively regulated by axons to generate non-myelin-forming and myelin-forming Schwann cells, respectively. Mol Cell Neurosci 8:336-50.

Lehming, N., Thanos, D., Brickman, J. M., Ma, J., Maniatis, T., and Ptashne, M. (1994). AnHMG-like protein that can switch a transcriptional activator to a repressor. Nature 371:175-9.

Lemke, G., and Chao, M. (1988). Axons regulate Schwann cell expression of the major myelinand NGF receptor genes. Development 102: 499-504.

Lemke, G., Lamar, E., and Patterson, J. (1988). Isolation and analysis of the gene encodingperipheral myelin protein zero. Neuron 1: 73-83.

Li, P., He, X., Gerrero, M. R., Mok, M., Aggarwal, A., and Rosenfeld, M. G. (1993). Spacing andorientation of bipartite DNA-binding motifs as potential functional determinants for POUdomain factors. Genes Dev 7: 2483-96.

Li, X., Wrabetz, L., Cheng, Y., and Kamholz, J. (1994). A novel cyclic AMP response element,CACTTGATC, mediates forskolin induction of the myelin basic protein promoter in therat Schwannoma line, D6P2T. J Neurochem 63: 28-40.

Lobsiger, C. S., Taylor, V., and Suter, U. (2002). The early life of a Schwann cell. Biol Chem 383:245-53.

Lupski, J. R., de Oca Luna, R. M., Slaugenhaupt, S., Pentao, L., Guzzetta, V., Trask, B. J.,Saucedo-Cardenas, O., Barker, D. F., Killian, J. M., Garcia, C. A., Chakravarti, A., andPatel, P. I. (1991). DNA duplication associated with Charcot-Marie-Tooth disease type 1A.Cell 66: 219-232.

Magnaghi, V., Cavarretta, I., Galbiati, M., Martini, L., and Melcangi, R. C. (2001). Neuroactivesteroids and peripheral myelin proteins. Brain Res Brain Res Rev 37: 360-71.

REFERENCES

104

Magyar, J. P., Martini, R., Ruelicke, T., Aguzzi, A., Adlkofer, K., Dembic, Z., Zielasek, J., Toyka,K. V., and Suter, U. (1996). Impaired differentiation of Schwann cells in transgenic micewith increased PMP22 gene dosage. J Neurosci 16: 5351-60.

Maier, M., Berger, P., Nave, K. A., and Suter, U. (2002a). Identification of the Regulatory Regionof the Peripheral Myelin Protein 22 (PMP22) Gene That Directs Temporal and SpatialExpression in Development and Regeneration of Peripheral Nerves. Mol Cell Neurosci 20:93-109.

Maier, M., Berger, P., and Suter, U. (2002b). Understanding Schwann cell-neurone interactions:the key to Charcot- Marie-Tooth disease? J Anat 200: 357-66.

Mandemakers, W., Zwart, R., Jaegle, M., Walbeehm, E., Visser, P., Grosveld, F., and Meijer, D.(2000). A distal Schwann cell-specific enhancer mediates axonal regulation of the Oct-6transcription factor during peripheral nerve development and regeneration. Embo J 19:2992-3003.

Manfioletti, G., Ruaro, M. E., Del Sal, G., Philipson, L., and Schneider, C. (1990). A growtharrest-specific (gas) gene codes for a membrane protein. Mol Cell Biol 10: 2924-30.

Mansuy, I. M., and Suter, U. (2000). Mouse genetics in cell biology. Exp Physiol 85: 661-79.

Martini, R., Schachner, M., and Brushart, T. M. (1994). The L2/HNK-1 carbohydrate ispreferentially expressed by previously motor axon-associated Schwann cells inreinnervated peripheral nerves. J Neurosci 14: 7180-91.

Maycox, P. R., Ortuno, D., Burrola, P., Kuhn, R., Bieri, P. L., Arrezo, J. C., and Lemke, G.(1997). A transgenic mouse model for human hereditary neuropathy with liability topressure palsies. Mol Cell Neurosci 8: 405-16.

Mayor, C., Brudno, M., Schwartz, J. R., Poliakov, A., Rubin, E. M., Frazer, K. A., Pachter, L. S.,and Dubchak, I. (2000). VISTA : visualizing global DNA sequence alignments of arbitrarylength. Bioinformatics 16: 1046-7.

Meech, R., Kallunki, P., Edelman, G. M., and Jones, F. S. (1999). A binding site forhomeodomain and Pax proteins is necessary for L1 cell adhesion molecule geneexpression by Pax-6 and bone morphogenetic proteins. Proc Natl Acad Sci U S A 96:2420-5.

Melcangi, R. C., Magnaghi, V., Cavarretta, I., Zucchi, I., Bovolin, P., D'Urso, D., and Martini, L.(1999). Progesterone derivatives are able to influence peripheral myelin protein 22 and P0gene expression: possible mechanisms of action. J Neurosci Res 56: 349-57.

Melcangi, R. C., Magnaghi, V., Galbiati, M., and Martini, L. (2001). Steroid effects on the geneexpression of peripheral myelin proteins. Horm Behav 40: 210-4.

Messing, A., Behringer, R. R., Hammang, J. P., Palmiter, R. D., Brinster, R. L., and Lemke, G.(1992). P0 promoter directs expression of reporter and toxin genes to Schwann cells oftransgenic mice. Neuron 8: 507-20.

Milot, E., Strouboulis, J., Trimborn, T., Wijgerde, M., de Boer, E., Langeveld, A., Tan-Un, K.,Vergeer, W., Yannoutsos, N., Grosveld, F., and Fraser, P. (1996). Heterochromatin effectson the frequency and duration of LCR-mediated gene transcription. Cell 87: 105-14.

REFERENCES

105

Mirsky, R., and Jessen, K. R. (1999). The neurobiology of Schwann cells. Brain Pathol 9: 293-311.

Mirsky, R., Jessen, K. R., Brennan, A., Parkinson, D., Dong, Z., Meier, C., Parmantier, E., andLawson, D. (2002). Schwann cells as regulators of nerve development. J Physiol Paris 96:17-24.

Misra, R. P., Bronson, S. K., Xiao, Q., Garrison, W., Li, J., Zhao, R., and Duncan, S. A. (2001).Generation of single-copy transgenic mouse embryos directly from ES cells by tetraploidembryo complementation. BMC Biotechnol 1: 12.

Miura, M., Tamura, T., Aoyama, A., and Mikoshiba, K. (1989). The promoter elements of themouse myelin basic protein gene function efficiently in NG108-15 neuronal/glial cells.Gene 75: 31-8.

Monuki, E. S., Kuhn, R., and Lemke, G. (1993). Repression of the myelin P0 gene by the POUtranscription factor SCIP. Mech Dev 42: 15-32.

Monuki, E. S., Kuhn, R., Weinmaster, G., Trapp, B. D., and Lemke, G. (1990). Expression andactivity of the POU transcription factor SCIP. Science 249: 1300-3.

Nadon, N. L., Arnheiter, H., and Hudson, L. D. (1994). A combination of PLP and DM20transgenes promotes partial myelination in the jimpy mouse. J Neurochem 63: 822-33.

Naef, R., and Suter, U. (1999). Impaired intracellular trafficking is a common disease mechanismof PMP22 point mutations in peripheral neuropathies. Neurobiol Dis 6: 1-14.

Naef, R., and Suter, U. (1998). Many facets of the peripheral myelin protein PMP22 inmyelination and disease. Microsc Res Tech 41: 359-71.

Nagarajan, R., Svaren, J., Le, N., Araki, T., Watson, M., and Milbrandt, J. (2001). EGR2mutations in inherited neuropathies dominant-negatively inhibit myelin gene expression.Neuron 30: 355-68.

Nave, K. A., and Lemke, G. (1991). Induction of the myelin proteolipid protein (PLP) gene in C6glioblastoma cells: functional analysis of the PLP promotor. J Neurosci 11: 3060-9.

Nelson, C. C., Hendy, S. C., Shukin, R. J., Cheng, H., Bruchovsky, N., Koop, B. F., and Rennie,P. S. (1999). Determinants of DNA sequence specificity of the androgen, progesterone,and glucocorticoid receptors: evidence for differential steroid receptor response elements.Mol Endocrinol 13: 2090-107.

Neuberg, D. H., Sancho, S., and Suter, U. (1999). Altered molecular architecture of peripheralnerves in mice lacking the peripheral myelin protein 22 or connexin32. J Neurosci Res 58:612-23.

Paca-Uccaralertkun, S., Zhao, L. J., Adya, N., Cross, J. V., Cullen, B. R., Boros, I. M., and Giam,C. Z. (1994). In vitro selection of DNA elements highly responsive to the human T- celllymphotropic virus type I transcriptional activator, Tax. Mol Cell Biol 14: 456-62.

Palmer, C. N., Hsu, M. H., Griffin, H. J., and Johnson, E. F. (1995). Novel sequence determinantsin peroxisome proliferator signaling. J Biol Chem 270: 16114-21.

Paratore, C., Goerich, D. E., Suter, U., Wegner, M., and Sommer, L. (2001). Survival and glialfate acquisition of neural crest cells are regulated by an interplay between the transcriptionfactor Sox10 and extrinsic combinatorial signaling. Development 128: 3949-61.

REFERENCES

106

Pareek, S., Suter, U., Snipes, G. J., Welcher, A. A., Shooter, E. M., and Murphy, R. A. (1993).Detection and processing of peripheral myelin protein PMP22 in cultured Schwann cells.J. Biol. Chem. 268: 10372-10379.

Pareek, S., Notterpek, L., Snipes, G. J., Naef, R., Sossin, W., Laliberte, J., Iacampo, S., Suter, U.,Shooter, E. M., and Murphy, R. A. (1997). Neurons promote the translocation ofperipheral myelin protein 22 into myelin. J Neurosci 17: 7754-62.

Parmantier, E., Braun, C., Thomas, J. L., Peyron, F., Martinez, S., and Zalc, B. (1997). PMP-22expression in the central nervous system of the embryonic mouse defines potentialtransverse segments and longitudinal columns. J Comp Neurol 378: 159-72.

Parmantier, E., Cabon, F., Braun, C., D'Urso, D., Muller, H. W., and Zalc, B. (1995). Peripheralmyelin protein-22 is expressed in rat and mouse brain and spinal cord motoneurons. Eur JNeurosci 7: 1080-8.

Patel, P. I., Roa, B. B., Welcher, A. A., Schoener-Scott, R., Trask, B. J., Pentao, L., Snipes, G. J.,Garcia, C. A., Francke, U., Shooter, E. M., and et al. (1992). The gene for the peripheralmyelin protein PMP-22 is a candidate for Charcot-Marie-Tooth disease type 1A. NatGenet 1: 159-65.

Peirano, R. I., Goerich, D. E., Riethmacher, D., and Wegner, M. (2000). Protein zero geneexpression is regulated by the glial transcription factor Sox10. Mol Cell Biol 20: 3198-209.

Pribyl, T. M., Campagnoni, C. W., Kampf, K., Kashima, T., Handley, V. W., McMahon, J., andCampagnoni, A. T. (1993). The human myelin basic protein gene is included within a 179-kilobase transcription unit: expression in the immune and central nervous systems. ProcNatl Acad Sci U S A 90: 10695-9.

Quandt, K., Frech, K., Karas, H., Wingender, E., and Werner, T. (1995). MatInd andMatInspector: new fast and versatile tools for detection of consensus matches innucleotide sequence data. Nucleic Acids Res 23: 4878-84.

Readhead, C., Schneider, A., Griffiths, I., and Nave, K. A. (1994). Premature arrest of myelinformation in transgenic mice with increased proteolipid protein gene dosage. Neuron 12:583-95.

Renner, K., Sock, E., Bermingham, J. R., Jr., and Wegner, M. (1996). Expression of thegene for the POU domain transcription factor Tst- 1/Oct6 is regulated by anestrogen-dependent enhancer. Nucleic Acids Res 24: 4552-7.

Roa, B. B., Garcia, C. A., Suter, U., Kulpa, D. A., Wise, C. A., Mueller, J., Welcher, A. A., Snipes,G. J., Shooter, E. M., Patel, P. I., and et al. (1993). Charcot-Marie-Tooth disease type 1A.Association with a spontaneous point mutation in the PMP22 gene. N Engl J Med 329: 96-101.

Rossant, J., Zirngibl, R., Cado, D., Shago, M., and Giguere, V. (1991). Expression of a retinoicacid response element-hsplacZ transgene defines specific domains of transcriptionalactivity during mouse embryogenesis. Genes Dev 5: 1333-44.

Saberan-Djoneidi, D., Sanguedolce, V., Assouline, Z., Levy, N., Passage, E., and Fontes, M.(2000). Molecular dissection of the Schwann cell specific promoter of the PMP22 gene.Gene 248: 223-31.

REFERENCES

107

Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). Molecular cloning: A laboratory manual.,Second Edition Edition: Cold Spring Harbor Laboratory Press).

Sancho, S., Magyar, J. P., Aguzzi, A., and Suter, U. (1999). Distal axonopathy in peripheralnerves of PMP22-mutant mice. Brain 122: 1563-77.

Sasaki, Y., Ishida, S., Morimoto, I., Yamashita, T., Kojima, T., Kihara, C., Tanaka, T., Imai, K.,Nakamura, Y., and Tokino, T. (2002). The p53 family member genes are involved in theNotch signal pathway. J Biol Chem 277: 719-24.

Scherer, S. S. (1997). The biology and pathobiology of Schwann cells. Curr Opin Neurol 10:386-97.

Scherer, S. S. (1999). Nodes, paranodes, and incisures: from form to function. Ann N Y Acad Sci883: 131-42.

Scherer, S. S., and Arroyo, E. J. (2002). Recent progress on the molecular organization ofmyelinated axons. J Peripher Nerv Syst 7: 1-12.

Scherer, S. S., Vogelbacker, H. H., and Kamholz, J. (1992). Axons modulate the expression ofproteolipid protein in the CNS. J Neurosci Res 32: 138-48.

Schneider, C., King, R. M., and Philipson, L. (1988). Genes specifically expressed at growtharrest of mammalian cells. Cell 54: 787-93.

Schreiber, J., Enderich, J., Sock, E., Schmidt, C., Richter-Landsberg, C., and Wegner, M. (1997).Redundancy of class III POU proteins in the oligodendrocyte lineage. J Biol Chem 272:32286-93.

Sereda, M., Griffiths, I., Puhlhofer, A., Stewart, H., Rossner, M. J., Zimmermann, F., Magyar, J.P., Schneider, A., Hund, E., Meinck, H.-M., Suter, U., and Nave, K. A. (1996). A rattransgenic model for Charcot-Marie-Tooth disease. Neuron 16: 1049-1060.

Sim, F., Zhao, C., Li, W., Lakatos, A., and Franklin, R. (2002). Expression of the POU-DomainTranscription Factors SCIP/Oct-6 and Brn-2 Is Associated with Schwann Cell but NotOligodendrocyte Remyelination of the CNS. Mol Cell Neurosci 20: 669.

Skre, H. (1974). Genetic and clinical aspects of Charcot-Marie-Tooth's disease. Clin Genet 6: 98-118.

Smit, A. F. (1996). The origin of interspersed repeats in the human genome. Curr Opin GenetDev 6: 743-8.

Snipes, G. J., Suter, U., and Shooter, E. M. (1993). Human peripheral myelin protein-22 carriesthe L2/HNK-1 carbohydrate adhesion epitope. J. Neurochem. 61: 1961-1964.

Snipes, G. J., Suter, U., Welcher, A. A., and Shooter, E. M. (1992). Characterization of a novelperipheral nervous system myelin protein (PMP-22/SR13). J Cell Biol 117: 225-38.

Spreyer, P., Kuhn, G., Hanemann, C. O., Gillen, C., Schaal, H., Kuhn, R., Lemke, G., and Muller,H. W. (1991). Axon-regulated expression of a Schwann cell transcript that is homologousto a 'growth arrest-specific' gene. Embo J 10: 3661-8.

Suter, U., Moskow, J. J., Welcher, A. A., Snipes, G. J., Kosaras, B., Sidman, R. L., Buchberg, A.M., and Shooter, E. M. (1992a). A leucine-to-proline mutation in the putative first

REFERENCES

108

transmembrane domain of the 22-kDa peripheral myelin protein in the trembler-J mouse.Proc Natl Acad Sci U S A 89: 4382-6.

Suter, U., Snipes, G. J., Schoener-Scott, R., Welcher, A. A., Pareek, S., Lupski, J. R., Murphy, R.A., Shooter, E. M., and Patel, P. I. (1994). Regulation of tissue-specific expression ofalternative peripheral myelin protein-22 (PMP22) gene transcripts by two promoters. JBiol Chem 269: 25795-808.

Suter, U., Welcher, A. A., Ozcelik, T., Snipes, G. J., Kosaras, B., Francke, U., Billings, G. S.,Sidman, R. L., and Shooter, E. M. (1992b). Trembler mouse carries a point mutation in amyelin gene. Nature 356: 241-244.

Suter, U., and Snipes, G. J. (1995). Biology and genetics of hereditary motor and sensoryneuropathies. Annu Rev Neurosci 18: 45-75.

Sutherland, H. G., Martin, D. I., and Whitelaw, E. (1997). A globin enhancer acts by increasingthe proportion of erythrocytes expressing a linked transgene. Mol Cell Biol 17: 1607-14.

Swirnoff, A. H., and Milbrandt, J. (1995). DNA-binding specificity of NGFI-A and related zincfinger transcription factors. Mol Cell Biol 15: 2275-87.

Taillebourg, E., Buart, S., and Charnay, P. (2002). Conditional, floxed allele of the Krox20 gene.Genesis 32: 112-3.

Tamura, T., Aoyama, A., Inoue, T., Miura, M., Okano, H., and Mikoshiba, K. (1989). Tissue-specific in vitro transcription from the mouse myelin basic protein promoter. Mol Cell Biol9: 3122-6.

Tamura, T., Miura, M., Ikenaka, K., and Mikoshiba, K. (1988). Analysis of transcription controlelements of the mouse myelin basic protein gene in HeLa cell extracts: demonstration of astrong NFI- binding motif in the upstream region. Nucleic Acids Res 16: 11441-59.

Tamura, T., Sumita, K., Hirose, S., and Mikoshiba, K. (1990). Core promoter of the mousemyelin basic protein gene governs brain- specific transcription in vitro. Embo J 9: 3101-8.

Taveggia, C., Pizzagalli, A., Feltri, M. L., Grinspan, J. B., Kamholz, J., and Wrabetz, L. (1998).MEBA derepresses the proximal myelin basic protein promoter in oligodendrocytes. JBiol Chem 273: 27741-8.

Taylor, V., Welcher, A. A., Program, A. E., and Suter, U. (1995). Epithelial membrane protein-1,peripheral myelin protein 22, and lens membrane protein 20 define a novel gene family. JBiol Chem 270: 28824-33.

Timmerman, V., De Jonghe, P., Ceuterick, C., De Vriendt, E., Lofgren, A., Nelis, E., Warner, L.E., Lupski, J. R., Martin, J. J., and Van Broeckhoven, C. (1999). Novel missense mutationin the early growth response 2 gene associated with Dejerine-Sottas syndrome phenotype.Neurology 52: 1827-32.

Toews, A. D., Hostettler, J., Barrett, C., and Morell, P. (1997). Alterations in gene expressionassociated with primary demyelination and remyelination in the peripheral nervoussystem. Neurochem Res 22: 1271-80.

Topilko, P., Schneider-Maunoury, S., Levi, G., Baron-Van Evercooren, A., Chennoufi, A. B.,Seitanidou, T., Babinet, C., and Charnay, P. (1994). Krox-20 controls myelination in theperipheral nervous system. Nature 371: 796-9.

REFERENCES

109

Trapp, B. D., Hauer, P., and Lemke, G. (1988). Axonal regulation of myelin protein mRNA levelsin actively myelinating Schwann cells. J Neurosci 8: 3515-21.

Umemori, H., Kadowaki, Y., Hirosawa, K., Yoshida, Y., Hironaka, K., Okano, H., and Yamamoto,T. (1999). Stimulation of myelin basic protein gene transcription by Fyn tyrosine kinasefor myelination. J Neurosci 19: 1393-7.

van de Wetering, R. A., Gabreels-Festen, A. A., Kremer, H., Kalscheuer, V. M., Gabreels, F. J.,and Mariman, E. C. (1999). Regulation and expression of the murine PMP22 gene. MammGenome 10: 419-22.

Voyvodic, J. T. (1989). Target size regulates calibre and myelination of sympathetic axons.Nature 342: 430-3.

Wang, M. M., and Reed, R. R. (1993). Molecular cloning of the olfactory neuronal transcriptionfactor Olf-1 by genetic selection in yeast. Nature 364: 121-6.

Warner, L. E., Mancias, P., Butler, I. J., McDonald, C. M., Keppen, L., Koob, K. G., and Lupski,J. R. (1998). Mutations in the early growth response 2 (EGR2) gene are associated withhereditary myelinopathies. Nat Genet 18: 382-4.

Wegner, M. (2000a). Transcriptional control in myelinating glia: flavors and spices. Glia 31: 1-14.

Wegner, M. (2000b). Transcriptional control in myelinating glia: the basic recipe. Glia 29: 118-23.

Welcher, A. A., Suter, U., De Leon, M., Snipes, G. J., and Shooter, E. M. (1991). A myelinprotein is encoded by the homologue of a growth arrest- specific gene. Proc Natl Acad SciU S A 88: 7195-9.

Wight, P. A., Duchala, C. S., Readhead, C., and Macklin, W. B. (1993). A myelin proteolipidprotein-LacZ fusion protein is developmentally regulated and targeted to the myelinmembrane in transgenic mice. J Cell Biol 123: 443-54.

Windebank, A. J. (1993). Inherited recurrent focal neuropathies. In: Dyck PJ, Thomas PK, GrffinJW, Low PA, Poduslo JF, editors. Peripheral Neuropathy.

Wu, R., Jurek, M., Sundarababu, S., and Weinstein, D. E. (2001). The POU gene Brn-5 isinduced by neuregulin and is restricted to myelinating Schwann cells. Mol Cell Neurosci17: 683-95.

Yamada, H., Komiyama, A., and Suzuki, K. (1995). Schwann cell responses to forskolin andcyclic AMP analogues: comparative study of mouse and rat Schwann cells. Brain Res 681:97-104.

Ye, P., Laszkiewicz, I., Wiggins, R. C., and Konat, G. W. (1994). Transcriptional regulation ofmyelin associated glycoprotein gene expression by cyclic AMP. J Neurosci Res 37: 683-90.

Yool, D. A., Edgar, J. M., Montague, P., and Malcolm, S. (2000). The proteolipid protein geneand myelin disorders in man and animal models. Hum Mol Genet 9: 987-92.

Young, P., and Suter, U. (2001). Disease mechanisms and potential therapeutic strategies inCharcot- Marie-Tooth disease. Brain Res Brain Res Rev 36: 213-21.

REFERENCES

110

Zanazzi, G., Einheber, S., Westreich, R., Hannocks, M. J., Bedell-Hogan, D., Marchionni, M. A.,and Salzer, J. L. (2001). Glial growth factor/neuregulin inhibits Schwann cell myelinationand induces demyelination. J Cell Biol 152: 1289-99.

Zoidl, G., Blass-Kampmann, S., D'Urso, D., Schmalenbach, C., and Muller, H. W. (1995).Retroviral-mediated gene transfer of the peripheral myelin protein PMP22 in Schwanncells: modulation of cell growth. Embo J. 14: 1122-1128.

Zoidl, G., D'Urso, D., Blass-Kampmann, S., Schmalenbach, C., Kuhn, R., and Muller, H. W.(1997). Influence of elevated expression of rat wild-type PMP22 and its mutantPMP22Trembler on cell growth of NIH3T3 fibroblasts. Cell Tissue Res 287: 459-70.

Zorick, T. S., and Lemke, G. (1996). Schwann cell differentiation. Curr Opin Cell Biol 8: 870-6.

Zorick, T. S., Syroid, D. E., Brown, A., Gridley, T., and Lemke, G. (1999). Krox-20 controlsSCIP expression, cell cycle exit and susceptibility to apoptosis in developing myelinatingSchwann cells. Development 126: 1397-406.

ACKNOWLEDGMENTS

111

11 ACKNOWLEDGMENTS

I would like to thank:

Prof. Dr. Ueli Suter for giving me the opportunity to work in his group and for his

scientific and personal support over the years.

Prof. Dr. Peter Sonderegger for spending time reading my thesis.

Ned Mantei for his time and patience correcting my thesis, his availability in the lab with

a lot of good advice, and for all the bike t(r)ips.

Prof. Martin Schwab, Suzie Atanasoski, Philipp Berger, Sara Sancho, Verdon Taylor and

Lukas Sommer for all the helpful and rewarding discussions and advice.

Annick Bonnet, Axel Nieman, Pit Young, Sonja Bonneick, Francois Castagner, Dino

Leone, Christian Paratore, Yves Benninger, Johanna Buchstaller for being nice friends

and collegues, for numerous helps and interesting discussions.

All the collegues of the Suter lab and of the IZB institute for the stimulating and nice

working atmosphere.

Last but not least, I thank my parents and Carina for their continous help and endless

support.

CURRICULUM VITAE

112

12 CURRICULUM VITAE

Name Marcel Maier Date of Birth 23rd May, 1974Citizenship SwissSex maleMarital status single

Home Address Alte Sulzbacherstr. 22, 8610 Uster, SwitzerlandPhone: +41 (0)1 940 70 78

Present Address Buchfinkenstrasse 6, CH-8052 Zürich, SwitzerlandPhone: +41 (0)1 301 08 58, E-mail: [email protected]

Education and Awards

1981-1987 Primary School, Uster, Switzerland1987-1993 High School, Kantonsschule Zürcher Oberland, Wetzikon, Switzerland

Degree obtained: Matura type C (natural science)

1991 Award of the Swiss Organisation "Science and Youth" for a thesis inthe area of ecological investigation of dragonflies

1993-1998 Study of biology at the Swiss Federal Institute of Technology (ETH)Zürich, Switzerland ; Courses and examinations in: math, chemistry,physics, biochemistry, molecular biology, physiology, neurobiologyand cell biology

May 1998 Diploma, Dipl. Natw. ETH in cell biology, immunology, molecularbiology, biochemistry and genetics at the Swiss Federal Institute ofTechnology (ETH) Zürich, Switzerland

1999-2002 Ph.D. student, Institute of Cell Biology, ETH Zürich, SwitzerlandArea of research: Molecular and Cellular NeurobiologyTitle of the Project: Transcriptional regulation of the CMT1A-diseasegene Peripheral Myelin Protein (PMP22)

2002 ASN Young Investigator Education Enhancement Award for theAmerican Society of Neurochemistry (ASN) Meeting

Scientific Training and Research experience

1999-2002 Postgraduate courses and exams in different areas of neurobiology atthe Neuroscience Center Zürich, University of Zürich, Switzerland

1999-2002 Ph. D. training in laboratory techniques and methods:

CURRICULUM VITAE

113

molecular biology: standard DNA cloning techniques, RNAtechniques, DNA extractions, Northern and Southern blot, quantitativereal time (RT)-PCR (TaqMan), In situ RNA Hybridisationcell biology: immunohistochemistry, classical histological stainings,semithin sections, electron microscopycell culture: standard cell culture techniques, primary mouse Schwanncell cultures, DRG explant cultures/in vitro myelination,cotransfection studies & reporter gene assays, adenoviral infectionsanimal experiments: establishing and maintaining transgenic mouseand rat lines, performing sciatic nerve crushes

2001 Training in laboratory animal experiments (LTK Module 1) at theInstitute of Laboratory Animal Sciences, University of Zürich,Switzerland: Handling and treatment of animals, welfare and ethicalissues

Additional professional skills and acitivites

1993-1996 Leader of Boy Scout troop with ca. 50 members

September Training in clinical chemistry and analysis at the Hospital - October 1996 Lachen, Switzerland during military service.

December 1998 General assistant in the Management unit of the information - February 1999 technology department at Union Bank of Switzerland (UBS),

Zürich

Languages

German: mother tongueEnglish: fluent written and spoken; 5 years during high school; First Certificate

1993; 1996: 1 month English language course in San Diego, USASpanish: intermediate; 1 year during high school, 1998: 1 month Spanish

language course in Couzco, Peru French: intermediate; 6 years during High School

Conferences and presentations

D-BIOL Symposium ETHZ 2000 and 2002, Davos, Switzerland. Maier M., Berger P.and Suter U. Promoter analysis of the CMT1A- disease gene PMP22 in vivo.

American Society for Neurochemistry, 33rd annual meeting, 2002. Florida, USASelection for oral presentation: Maier M., Berger P., Nave K.-A. and Suter U. Promoteranalysis of the CMT1A- disease gene PMP22 in vivo. (J Neurochem. 2002 Jun;81(Suppl. 1):77)

CURRICULUM VITAE

114

Publications

Maier M., Wildermuth H. (1991), Oekologische Beobachtungen zur Emergenz einigerAnisopteren an Kleingewässer. Libellula 10 (3/4): 89-104

Maier M., Berger P., Nave K.-A and Suter U.(2002), Identification of the RegulatoryRegion of the Peripheral Myelin Protein 22 (PMP22) Gene that Directs Temporal andSpatial Expression in Development and Regeneration of Peripheral Nerves. Mol. CellNeurosci (2002); 20(1):93-109

Maier M., Berger P. and Suter U. (2002), Understanding Schwann Cell-NeuronInteractions: The Key to Charcot-Marie-Tooth Disease ? J Anat (2002), 200; 357-366

Maier M., Castagner F., Berger P. and Suter U. (2003). Dissection of the PeripheralMyelin Protein 22 (PMP22) Promoter in vivo Reveals a Late Myelinating Schwann CellSpecific Element, in preparation

Referees

Prof. Ueli Suter, Ph. D., Head of Institute for Cell Biology, HPM E39, ETH Hönggerberg, CH-8093 Zürich, Switzerland, Phone: +41 1 633 34 32; Fax: +41 1 633 11 90, e-mail: [email protected]

Prof. Peter Sonderegger, Department of Biochemistry, University of Zürich, Winterthurerstr. 190, CH-8057 Zürich, Switzerland, Phone: +41 1 635 55 41; Fax: +41 1 635 68 31e-mail: [email protected]

rights / license: research collection in copyright - non ...26200/et… · promoter analysis and...

Documents